Post Modification of Liquid Polybutadienes and Their ...

Post Modification of Liquid Polybutadienes

and Their Rheological Properties

Dissertation

with the aim of achieving the doctoral degree at the Faculty of Mathematics, Informatics and

Natural Sciences

submitted to the

Department of Chemistry

University of Hamburg

by

Hannes Jürgens

Hamburg 2018

Date of oral defense: 22.02.2019

Approval for printing: 22.02.2019

The experimental work described in this thesis has been carried out between May 2012 and July

2016 at the Institute of Technical and Macromolecular Chemistry, University of Hamburg in

the research group of Professor Dr. Gerrit A. Luinstra

The following evaluators recommend the admission of the dissertation:

1. Evaluator: Professor Dr. Gerrit A. Luinstra

2. Evaluator: Professor Dr. Hans-Ulrich Moritz

I

Table of Contents

List of Abbreviations .................................................................................................................. 1

1 Abstract .............................................................................................................................. 5

2 Zusammenfassung .............................................................................................................. 7

3 Introduction and Background ............................................................................................. 9

3.1 Synthetic Rubber ........................................................................................................ 9

3.2 Polybutadiene ........................................................................................................... 10

3.2.1 Introduction .......................................................................................................... 10

3.2.2 Anionic Polymerization of Polybutadiene ........................................................... 11

3.2.3 Radical Polymerization of Polybutadiene ............................................................ 12

3.2.4 Coordinative Polymerization of Polybutadiene ................................................... 12

3.2.5 Liquid Polybutadiene ........................................................................................... 14

3.2.6 Post Modification of Polybutadiene ..................................................................... 15

3.3 Rheology and Viscoelasticity ................................................................................... 18

3.3.1 Cox-Merz Relation ............................................................................................... 20

3.3.2 Small Amplitude Oscillatory Shear (SAOS) ........................................................ 20

3.3.3 Time Temperature Superposition ......................................................................... 21

4 Motivation ........................................................................................................................ 23

5 Results and Discussion ..................................................................................................... 24

5.1 Scalable Semi-Batch Process for Epoxidation of Polybutadienes ........................... 24

5.1.1 Epoxidation with Hydrogen Peroxide Feed ......................................................... 29

5.1.2 Larger Scale Reactions ......................................................................................... 36

5.1.3 Higher molecular weight polybutadienes ............................................................. 37

5.1.4 Material Properties ............................................................................................... 39

5.1.5 Summary .............................................................................................................. 42

5.2 Rheological Properties of Epoxidized Liquid Rubber in Bulk and Solution ........... 43

II

5.2.1 Molecular Weight Dependance ............................................................................ 44

5.2.2 Concentration Dependance .................................................................................. 45

5.2.3 Functionality dependence ..................................................................................... 46

5.2.4 Interim Conclusion ............................................................................................... 47

5.2.5 Temperature Dependence ..................................................................................... 48

5.2.6 Conclusion ............................................................................................................ 50

5.3 Aminolysis of Epoxidized Polybutadienes Catalyzed by Lithium Bromide ............ 52

5.3.1 Reaction Mechanism ............................................................................................ 52

5.3.2 Analysis of a Model System ................................................................................. 52

5.3.3 Progression of Aminolysis with Butyl Amine ..................................................... 54

5.3.4 Intramolecular Cyclization ................................................................................... 58

5.3.5 Higher Molecular Mass Epoxidized Polybutadienes ........................................... 62

5.3.6 Material Properties ............................................................................................... 62

5.3.7 Summary .............................................................................................................. 64

5.4 Quaternization of aminated polybutadienes ............................................................. 66

5.4.1 Reaction Mechanism ............................................................................................ 66

5.4.2 Analysis of a Model System ................................................................................. 66

5.4.3 Progression of Quaternization with Iodomethane ................................................ 68

5.4.4 Material properties ............................................................................................... 70

6 Summary .......................................................................................................................... 72

7 Experimental Part ............................................................................................................. 74

7.1 Materials ................................................................................................................... 74

7.2 Nuclear Magnetic Resonance (NMR) ...................................................................... 74

7.3 Matrix Assisted Laser Desorption (Time of Flight) Mass Spectrometry (MALDI) 74

7.4 Rheometry ................................................................................................................ 75

7.5 Differential Scanning Calorimetry (DSC) ................................................................ 75

7.6 Gel Permeation Chromatography (GPC) ................................................................. 75

7.7 Titration of Epoxidized Polybutadienes ................................................................... 75

III

7.8 Epoxidation of Polybutadienes ................................................................................. 76

7.8.1 Reaction Setup ...................................................................................................... 76

7.8.2 General Procedure of Epoxidation with in-situ Formed Performic Acid ............. 76

7.8.3 Degree of Epoxidation ......................................................................................... 77

7.8.4 Titration of Epoxide Groups ................................................................................ 78

7.9 Rheological Characterization of Epoxidized Polybutadienes .................................. 79

7.9.1 Materials ............................................................................................................... 79

7.10 Oscillatory Rheology measurements ........................................................................ 79

7.11 Aminolysis of Epoxidized Polybutadienes .............................................................. 80

7.11.1 Process .............................................................................................................. 80

7.11.2 Degree of Aminolysis ....................................................................................... 80

7.12 Quaternization of aminated polybutadienes ............................................................. 84

7.12.1 Process .............................................................................................................. 84

7.12.2 Degree of Quaternization ................................................................................. 84

8 Safety Data ....................................................................................................................... 87

9 Bibliography ..................................................................................................................... 91

10 Acknowledgements .......................................................................................................... 99

11 Declaration of Oath ........................................................................................................ 101

List of Abbreviations

1


|η*| Complex Oscillatory Viscosity

ABR Aminated Butadiene Rubber

aq. Aqueous

bp Boiling Point

BR Butadiene Rubber

c Concentration

cat. Catalyst

CR Chloroprene Rubber

DSC Dynamic Scanning Calorimetry

EA Activation Energy

EBR Epoxidated Butadiene Rubber

EP-value Epoxidation Value (Titration)

EPDM Ethylene Propylene Diene Rubber

eq. Equivalents

E-SBR Emulsion Styrene Butadiene Rubber

f Epoxidation Ratio

FA Formic Acid

g Structural Factor

G’ Oscillatory Shear Storage Modulus


2

G’’ Oscillatory Shear Loss Modulus

GPC Gel Permeation Chromatography

HP Hydrogen Peroxide

I-Gated Inverse Gated Decoupling Experiment

IIR Isobutylene Isoprene Rubber

IR Isoprene Rubber

k reaction rate constant

LiBr Lithium Bromide

M Molecular Weight

m Mass

MALDI-Tof Matrix Assisted Laser Desorption Ionisation – Time of Flight

Mn Number Average Molecular Weight

Mw Mass Average Molecular Weight

NBR Acrylonitrile Butadiene Rubber

NMR Nuclear Magnetic Resonance

PBu Polybutadiene

PFA Performic Acid

ppm Parts per Million

PTFE Polytetrafluorethylene (Teflon®)

R Universal Gas Constant

SBR Styrene-Butadiene-Rubber

SBC Styrenic Block Copolymer Thermoplastic Elastomer


3

SEC Size Exclusion Chromatography

SN2 Bimolecular Nucleophilic Substitution

S-SBR Solution Styrene Butadiene Rubber

T Temperature

T0 Reference Temperature

Tg Glass Transition Temperature

Tm Melting Point

THF Tetrahydrofurane

tKOH Titer

TTS Time Temperature Superposition

V Volume

w Weight Fraction

X Conversion

αT Shift Factor

δ Chemical Shift

η0 Zero Shear Viscosity

ξ Fiction Factor

φ Volume Fraction

ω Angular Frequency


4

Abstract

5

1 Abstract

Polymers carrying functional groups for specific applications are accessible through two

different routes. The polymerization of monomers showing the desired functionality or the

post modification with the functional groups of existing polymers.[1] The former route

requires a polymerization environment that is tolerating the functional group. The process

often has limited versatility, because a monomer specific process is used. Post modification

on the other hand poses additional process steps increasing the production costs, especially

if expensive catalysts have to be used. In this thesis a versatile post modification process is

shown with easy accessible catalytic systems for the functionalization of polybutadienes.

The modification was done using low vinyl polybutadienes of molecular weights between

1500 and 26000 g/mol.

A two-phase semi batch epoxidation process using in-situ formed performic acid gave access

to a versatile intermediate which could be further functionalized. Continuous addition of

aqueous hydrogen peroxide to a toluene solution of polybutadiene, catalyzed by formic acid

gave a controllable, safe and scalable reaction. The reaction progress could be monitored by 1H-NMR spectroscopy or titration. The kinetics of the reaction were investigated and the

formation of performic acid was identified as rate determining step. The activation energy

using an Arrhenius-plot was determined and its value of 29 kJ/mol is in the range of those,

shown in the literature.[2,3]

The glass transition temperature of the epoxidated polybutadienes rose with increasing

epoxide content following the fox-equation. The bulk viscosity increased about one order of

magnitude by epoxidation of about 50%.

The versatile epoxidized polybutadienes described above were characterized by oscillatory

rheology. The change in viscosity by introduction of epoxide groups makes it necessary to

predict the product properties for optimal process design. Using the model introduced here

the final viscosity can be calculated based on conversion, molecular weight of

polybutadiene, concentration in toluene and reaction temperature.

The epoxide groups on the polymer backbone can be opened by nucleophiles of which

amines were investigated as an example in this thesis. A one step process with high

conversion using n-butylamine catalyzed by lithium bromide was developed. The kinetics

Abstract

6

of the reaction were looked at and a cyclization side reaction competing with the addition of

n-butylamine was identified.

The introduction of polar amine groups further increased the glass transition temperature to

about -20 °C at 30% amine content relative to initial double bond concentration. Signs of

conformational changes in the polymer chain could be observed using SEC measurements.

The secondary amine groups in aminated polybutadienes were methylated using

iodomethane and potassium carbonate to introduce ammonium ions. The reaction proceeded

at room temperature with an excess of iodomethane. The resulting polymer was a hard and

brittle material with a glass transition temperature of about 45 °C at 30% ammonium ion

content relative to initial double bond concentration. The solubility in polar solvents like

methanol increased significantly.

Zusammenfassung

7

2 Zusammenfassung

Polymere mit funktionellen Gruppen sind auf zwei verschiedenen Wegen zugänglich.

Funktionelle monomere können polymerisiert werden oder existierende Polymere können in

einer Post-Modifikation nachträglich funktionalisiert werden.[1] Die erste Route erfordert

eine Polymerisationsumgebung, welche die funktionelle Gruppe toleriert. Dieser Prozess ist

oft nur bedingt flexibel, da ein monomerspezifischer Prozess verwendet wird. Post-

Modifikation hat andererseits den Nachteil, dass zusätzliche Prozessschritte erforderlich

werden, was die Produktionskosten erhöht. Das gilt insbesondere, wenn teure

Katalysatorsysteme verwendet werden. In dieser Arbeit wird ein variabler Prozess gezeigt,

der die Post-Modifikation von Polybutadienen unter Verwendung von leicht zugänglichen

Katalysatorsystemen ermöglicht. Die Modifikationen wurden durchgeführt an

Polybutadienen mit niedrigem Vinylgehalt der Molmassen 1500 bis 26000 g/mol.

Ein zweiphasiger Semi-Batch-Prozess unter Verwendung von in-situ hergestellter

Perameisensäure machte ein variables Intermediat zugänglich, welches weiter

funktionalisiert werden konnte. Wässriges Wasserstoffperoxid wurde kontinuierlich zu einer

toluolischen Polybutadienlösung gegeben, welche Ameisensäure als Katalysator enthielt.

Dadurch wurde ein kontrollierbarer, sicherer und skalierbarer Prozess geschaffen. Der

Reaktionsfortschritt konnte per 1H-NMR und per Titration verfolgt werden. Die Kinetik der

Reaktion wurde untersucht und die Bildung der Perameisensäure als

geschwindigkeitsbestimmender Schritt identifiziert. Die Aktivierungsenergie wurde mit

Hilfe einer Arrheniusauftragung bestimmt. Der ermittelte Wert der Aktivierungsenergie von

29 kJ/mol liegt im Bereich der literaturbekannten Werte.[2,3]

Die Glassübergangstemperatur der epoxidierten Polybutadiene erhöhte sich mit steigendem

Epoxidgehalt und der Verlauf folgte der Fox-Gleichung. Die Viskosität erhöhte sich um etwa

eine Größenordnung bei einem Epoxidgehalt von 50%.

Die oben beschriebenen epoxidierten Polybutadiene wurden mit Hilfe von

Oszillationsmessungen rheologisch untersucht. Die Viskositätsänderung durch die

Einführung von Epoxidgruppen macht eine Abschätzung der Produkteigenschaften

unabdingbar für eine industrielle Prozessentwicklung. Das Modell, welches hier vorgestellt

wird, erlaubt es, die Viskosität des Produktes zu berechnen bei Kenntnis des Umsatzes, des

Zusammenfassung

8

Molekulargewichtes des eingesetzten Polybutadiens, dessen Konzentration in Toluol und

der Reaktionstemperatur.

Die Epoxidgruppen an dem Polymerbackbone können durch Nukleophile geöffnet werden.

In dieser Arbeit wurde die Reaktion mit Aminen als Beispiel gängiger Nukleophile genauer

untersucht. Ein einstufiger Prozess wird vorgestellt, welcher einen hohen Umsatz unter

Verwendung von n-Butylamin und Lithiumbromid als Katalysator verspricht. Die Kinetik

der Reaktion wurde betrachtet und eine Cyclisierung als Nebenreaktion identifiziert, welche

mit der n-Butylaminaddition konkurriert.

Die Einführung polarer Amingruppen erhöhte die Glassübergangstemperatur weiter auf

etwa -20 °C bei 30% Amingehalt bezogen auf die Konzentration an Doppelbindungen des

Ausgangspolybutadiens. Außerdem konnten Hinweise auf eine Konformationsänderung

durch GPC-Messungen beobachtet werden.

Die sekundären Amine der funktionalisierten Polybutadiene wurden methyliert unter

Verwendung von Iodmethan und Kaliumcarbonat. Mit einem Überschuss an Iodmethan lief

die Reaktion bei Raumtemperatur bis zu quartären Ammoniumionen ab. Das so erhaltene

Polymer war ein hartes sprödes Material mit einer Glassübergangstemperatur von etwa

45 °C bei einem Gehalt an quartären Ammoniumionen von 30% bezogen auf die

Doppelbindungskonzentration der eingesetzten Polybutadiene. Die Löslichkeit in polaren

Lösungsmitteln wie Methanol stieg signifikant.

Introduction and Background

9

3 Introduction and Background

3.1 Synthetic Rubber

Rubber is a material used in many industries like automotive, tires, medical applications and

everyday life. Apart from natural rubber, synthetic rubbers are a growing field with

increasing production capacities worldwide. Butadiene Rubber (BR) and butadiene

containing copolymers like styrene-butadiene-rubber (SBR) made over 75% of the global

rubber production capacity in 2015.[4]

Figure 1. Global synthetic rubber production capacity 2015[4]

Even though SBR still is the most important synthetic rubber, BR production capacity was

the fastest growing in recent years. Production capacity for BR was increased by roughly

2000 kt (66%) between 2010 and 2015. This was mostly attributed to increasing demand for

high performance tires due to European tire labeling. That makes BR or Polybutadiene an

essential synthetic rubber nowadays with the biggest application in tire industries e.g. for

low rolling resistance.[5]


10

3.2 Polybutadiene

3.2.1 Introduction

The alkali metal catalyzed polymerization of butadiene was first described in the year

1910.[6–8] It was commercialized with the brand name BUNA in the 1920ies.[9] Stereoregular

polybutadienes (PBu) were introduced after the advances of Ziegler-Nata-Catalysts in the

1950ies.[10] The most important polymerization method today is the Ziegler-Natta-process

in solution. The largest quantities of polybutadiene today are used in tire industry.[5]

The properties of polybutadiene depend on the polymer microstructure. The butadiene can

be inserted in 1,4 or 1,2 (≡ 3,4) addition. The 1,4 isomers can be present in cis or trans form.

Additionally, a cyclic group can be formed out of two butadiene units carrying a terminal

double bond (Figure 2):

Figure 2. Polybutadiene isomers

Industrial scale polymerization of butadiene follows one of three different processes.

Anionic polymerization, radical polymerization or coordinative insertion polymerization.[11]

The microstructure of the polybutadiene depends on the method used.

Table 1. Microstructure and glass transition temperatures of polybutadienes[5]

Catalyst cis-1,4-content trans-1,4-content 1,2-content Tg [°C]

Nd – BR 97% 2% 1% -109

Co – BR 95% 3% 2% -107

Ni – BR 96% 2% 2% -107

Ti – BR 92% 4% 4% -105


11

Li – BR medium medium 10-50% -931

E - BR <20% ~70% <20% -781

3.2.2 Anionic Polymerization of Polybutadiene

Anionic polymerization of elastomers was the first synthetic route in elastomer synthesis. It

was described 1910 by different authors using sodium metal.[7,12,6] This process was later

responsible for the trade name BUNA® which is still used by Arlanxeo and Trinseo today.

The process was too expensive, which was why it wasn’t successful commercially. That was

achieved in the 1950ies with the introduction of homogenous catalysis using butyl lithium

and sodium naphthalene (Figure 3).[5]

Figure 3. Anionic initiation of unsaturated carbohydrates

The anionic polymerization is initiated by lithium alkyls and is generating polybutadienes

with medium cis-content and 10-50% vinyl content (depending on solvent polarity). The

higher the solvents polarity, the higher the vinyl content. That is because polar solvents

dissociate carbanion and metal ion and inhibit the formation of metal ion monomer

aggregates. An ethyl-lithium initiated anionic polymerization of butadiene gives about 7%

1,2-vinyl content in hexane and about 91% in THF.[5]

1 Varies with microstructure


12

Figure 4. Solvent influence on polybutadiene microstructure

3.2.3 Radical Polymerization of Polybutadiene

Radical polymerization is conducted in aqueous emulsion polymerization at low

temperatures (T = 5 °C). Redox systems like hydroperoxide or iron (II) complexes are used

as initiators. The resulting micro structure is a low cis-content and low vinyl-content (both

<20%). The industrial significance of radical polymerized polybutadiene is limited.[5]

3.2.4 Coordinative Polymerization of Polybutadiene

Coordinative polymerization of butadiene gives access to stereoregular polybutadienes. The

microstructure consists usually of cis contents above 90% and differs with the catalyst

system used. The commercially available Ziegler-Natta-polybutadienes are synthesized

using titanium, cobalt, nickel or neodymium based catalysts.[5]

The Mechanism mostly accepted today was postulated by COSSÉE and ARLMAN.[13–17] The

metal complex gets activated e.g. by triethylaluminium in a first step including the migration

of the vacant reactive site. An olefin can then be inserted into the growing polymer chain via

a four-center transition state (Figure 5).


13

Figure 5. Mechanism of propene polymerization by Ziegler-Natta-Catalyst postulated by COSSEE and ARLMAN

1,3-Butadiene gets inserted almost exclusively in cis-1,4 position because of a lower energy

transition state of the cis-1,4-insertion (Figure 6).[18]

Figure 6. 1,3-Butadiene insertion by Ziegler-Natta Catalysts

The Polymerization of Butadiene with Ziegler-Natta Catalysts is done in solution with

monomer concentrations of 10-15% to avoid high viscosities. The Ziegler-Natta catalyzed

polymerization is the commercially most important process today for polybutadiene.[5]


14

High-cis-polybutadienes can crystallize and usually have melting points below room

temperature. Polybutadienes made by radical or anionic polymerization are amorphous.

Melting points and crystallization half-life at -20 °C highly correlate with cis-1,4-content

and are listed in Table 2.[5]

Table 2. Melting points and crystallization half-life of different types of polybutadiene

Type cis-1,4-content Tm [°C] crystallization half-life at -20 °C [min]

Nd-BR 97 -7 7

Co-BR 95 -11 40

Ni-BR 96 -10 25

Ti-BR 92 -23 360

Li-BR 38 amorph -

E-BR 14 amorph -

3.2.5 Liquid Polybutadiene

The viscosity of polymers is, apart from other factors like microstructure, dependent on their

molecular weight. Since this dependency is continuous there is no sharp differentiation

between a solid and liquid possible. Polybutadienes here are considered liquid at molecular

weights below 50000 g/mol (for low vinyl content), since their flowability at room

temperature then is easily visible to the naked eye. Since the term liquid polybutadiene is

vague the grades used in this thesis are named low molecular weight polybutadienes, which

all have a molecular weight below 50000 g/mol.

The synthesis of liquid polybutadienes mainly follows two methods: 1) a “telomerisation”

system where e.g. toluene not only acts as a solvent but also as a transmetallating and chain

transfer agent or 2) “living polymerization”. Anionic polymerization is used for both routes

with n-butyl lithium as iniator. In route 1) in addition to the initiator, a promoter is used and

the active chain end can be transferred to toluene to form benzyl lithium, which itself can

initiate new chains. This CH exchange reaction is responsible for the benzyl end group of

the vast majority of polymer chains made by this process. Lithene ultra® PM4 is made via

route 1, e.g. Lithene ultra® N4 5000 is made via route 2.[19]


15

Low molecular weight polybutadiene (PBu) is a relatively cheap and readily available

material. It represents a technically useful class of polymers with a variety of applications.

It is equally applied as (reactive) plasticizer[20,21], binder for varnish colors[22], adhesive[23]

and stabilizer[24].

3.2.6 Post Modification of Polybutadiene

Functionalized rubber is of high interest for some years now e.g. to improve the SBR filler

interaction in tire rubber compounds.[25] The range of applications for low molecular

polybutadienes can further be increased e.g. by introduction of epoxide groups on the

polymer backbone.[26–32] Epoxidation changes the polarity of the material and transforms it

into a reactive compatibilizer.[33–35] Epoxidized PBu could also function as a base polymer

for reactive functionalization processes, and as access to functionalized polyolefins, i.e. after

hydrogenation.[36,37] Epoxidation of PBu may be achieved by a variety of routes[36]; the

epoxidation with carboxylic peracids is one of the most convenient processes concerning the

availability of chemicals and experimental effort[38,39]. Latter may be provided as a

presynthesized peracid[39–45] or by an in-situ formation from a carboxylic acid and hydrogen

peroxide.[38,46,47,2,48–51,3] The in-situ procedure has the advantage of using smaller (final)

amounts of acid and has better economics. A lower acid concentration has a positive effect

on the rate of side reactions.[39,50] Formic acid as carboxylic acid has additional advantages

as it combines the function of carboxylic acid for providing performic acid with toluene

solubility with a pKs value that is low enough to catalyze the reaction. It is also used in this

study.

The biphasic system of Scheme 1, comprising an organic phase holding the PBu and an

aqueous phase with performic acid, is useful for preparing partly epoxidized PBu. The major

part of the performic acid is formed in water by an (acid catalyzed) esterification of formic

acid with hydrogen peroxide. It subsequently transfers into the toluene phase, where the

double bonds in PBu are epoxized and formic acid is regenerated. Formic acid is thus an

effective catalyst for the net oxidation of PBu with hydrogen peroxide.


16

Scheme 1. Biphasic epoxidation of polybutadiene by in-situ formed performic acid

Several side reactions may occur, in particular ring opening of the formed epoxide groups

by reaction with formic acid. Formation of hydroxyl-formyl functionalized polybutadiene is

favored at higher formic acid concentration and at elevated temperatures[50]. Hydroxyl

groups on epoxidized PBu may crosslink the polymer by reaction with further epoxy entities.

Low concentrations of formic acid are therefore favored. Hydrogen peroxide may also be

decomposed in a redox disproportionation reaction, a reaction that is also faster at higher

concentrations of hydrogen peroxide. It is most commonly added to the reactor in one batch

because of the ease of handling.[38,2,52] The downside of the procedure next to the lower

efficiency (oxygen formation) is the lower safety in comparison to a semi-batch procedure,

the smaller control over the extent of epoxidation and the risk of epoxide ring opening

reactions especially at elevated temperatures. Therefore, a robust procedure has been

developed allowing to adjust the epoxide content to a desired level, and to provide kg scale

of material. The protocol uses the principle depicted in Scheme 1 with low formic acid

concentrations and the semi-batch operation in hydrogen peroxide.

These epoxidized polybutadienes represent a reactive intermediate for further

functionalization reactions. The functionalization also opens new application possibilities


17

such as electrolytes[53] or gas separation[54]. The addition of amines is therefore an interesting

field of research in which many possible reaction pathways are reported[36].

A simple and straightforward method to add amines to the polymers backbone is the

nucleophilic ring opening of epoxides[36]. Epoxide groups in the polybutadiene backbone,

however, show a markedly low reactivity towards nucleophiles such as amines, compared

to asymmetrically substituted epoxides[55]. Various catalysts were proposed over time for the

ring opening of small molecular epoxides. Metal halides[56–58], metal triflates[59–62], solid

phase fixed catalysts[63] and metal free catalysts[64,65] were studied for their activity in

epoxide ring opening. Most of these studies however used asymmetric epoxides or

cyclohexene oxide, which show a considerably higher reactivity towards the ring opening

compared to epoxides on the polybutadiene backbone.

Very few studies cover the ring opening of epoxidized polybutadienes with amines and those

published show low conversions of epoxide groups in the backbone.[36] In this study lithium

bromide was used in a solvent free process for the nucleophilic ring opening of epoxides in

low molecular weight polybutadiene with amines. A simple synthetic pathway is presented

to quantitatively open the epoxide groups in the backbone of polybutadiene. Aditionally the

kinetics of the ring opening, side reactions and changes in properties of the functionalized

polybutadienes were studied.

Amines can be alkylated with alkylating agents like iodomethane or dimethyl sulfate in the

presence of a base.[66–69] The methylation, if allowed, is proceeding until ammonium ion

formation. Amines are very reactive towards alkylation because of their high nucleophilicity.

The Alkylation with iodomethane usually is proceeding at room temperature without any

additional catalysts. The quaternization of polybutadienes was reported before using

modified initiators in anionic polymerization[70] or post modification reactions[71]. These

examples include either intervention in the polymerization process or precious metal

catalysts. A straight forward reaction scheme is presented here to introduce ammonium ions

into the polybutadiene backbone using simple reaction steps without precious metal

catalysts.

Ammonium ion carrying polymers can be used as ion exchanger[72] or antistatics[73].


18

3.3 Rheology and Viscoelasticity

Rheology is the science of flow and deformation. The correlation between the deformation

ε of an ideal spring induced by a force F can be described by Hooke’s law:

= ∗

E in this case is a material constant describing its elasticity. This can be transferred to the

uniaxial deformation of an ideal solid where the force is replaced by the stress (force per

area).[74]

In ideal fluids a correlation can be drawn between stress and deformation with time. In shear

deformation these values are shear stress σ and shear rate :

= ∗

The constant of proportionality is the viscosity η, which is independent on the shear rate.

Real materials are neither ideal solids nor ideal fluids which makes the dynamic viscosity η

dependent on the shear rate . The observed behavior can be e.g. shear thinning or shear

thickening. Shear thinning is observed with most polymer melts, where entanglements are

influencing the behavior at higher shear rates. They show newtonian behavior at low shear

rates, at higher shear rates the loosening of entanglements predominates, and the viscosity

decreases. The opposite behavior (shear thickening) can be observed e.g. due to the

formation of strain induced crystallization (Figure 7).


19

Figure 7. Newtonian, shear thinning (pseudo plastic) and shear thickening (dilatant) behavior

The range of shear rates in everyday life applications is very broad. Typical shear rates of

some applications are depicted in Figure 8.[75]

Figure 8. Shear rate and some applications


20

3.3.1 Cox-Merz Relation

Usual plate-plate rheometers cover a shear rate range of 10-2 to 102 s-1. Capillary viscometers

cover shear rate ranges between 10-1 s-1 and 104 s-1. Combined both methods can cover a

broad range of shear rate but requires extensive analytical effort and material.[1] An easy

empirical method was introduced 1958 by W. P. COX AND E. H. MERZ.[76] They postulated

the dynamic viscosity η( ) at a constant shear rate equals the absolute complex viscosity

|η*(ω)| at constant angular frequency ω if = . With this relation a broad range of shear

rate can be covered by simple oscillatory shear experiments. This relation is valid for most

unfilled polymer solution and melts and it makes the oscillatory shear experiment a powerful

tool in analyzing polymer melts.[77]

3.3.2 Small Amplitude Oscillatory Shear (SAOS)

In oscillatory shear experiments the deformation is not continuously increasing but is applied

sinusoidal. This sinusoidal deformation γ generates a stress response σ which is also

sinusoidal as long as the amplitude does not exceed the linear viscoelastic region of the

material.

Figure 9. Applied amplitude and stress response in an oscillatory shear experiment

In ideal solids the response comes without delay and the phase angle δ is 0 °. Ideal fluids

respond with a phase angle of 90 °. Viscoelastic materials like polymer melts have phase

angles somewhere in between. This complex stress response G* can be divided into elastic


21

and viscous contribution to the stress response and are represented by the storage modulus

G’ and the loss modulus G’’ which are defined as follows:

= ∗

= ∗

∗ = + ∗ ′′ The absolute complex viscosity can be determined from the absolute complex stress

response and the angular frequency

| ∗| = | ∗|

The range of frequency in rheometers is limited to around 10-3 to 103 rad/s.[75] It is also very

time consuming to measure at low frequencies. With time-temperature superposition the

range of frequency can be expanded.

3.3.3 Time Temperature Superposition

The time temperature superposition (TTS) usually is applied to non-crosslinked, unfilled

materials. Additionally, the following requirements for the investigated materials need to be

met for the applied temperature range:[77]

The investigated materials mustn’t follow erratic change in structural character

The material mustn’t undergo strain crystallization

The glass transition temperature of the material should be significantly below the

temperature of measurement

The principle of TTS states a comparable influence on rheological properties for the time

frame of shear stress and the temperature. A decrease in temperature has a similar influence

on shear moduli like the increase in frequency. Several measurements in a certain range of

frequency at different temperatures can therefore be shifted to a master curve with an

expanded range of frequency. Frequency sweeps measured at higher temperatures compared

to the reference temperature are shifted to lower frequencies and vice versa (Figure 10).


22

Figure 10. Schematic shifting of temperature sweep measurements to a master curve

Motivation

23

4 Motivation

The aim of this work is the development of a reaction process for post modification of

polybutadienes. The idea is to use the cheap and readily available polybutadiene and

introduce polar groups to the polymer backbone to widen the range of application.

The identified pathway includes the epoxidation of double bonds to get a reactive

intermediate. This intermediate can be further functionalized with various nucleophiles of

which amines as an example are considered in detail. Additionally, the amine groups are

alkylated towards ammonium ions to widen the range of possible applications even further.

Since the functionalization is expected to have an influence on the polybutadiene properties,

the rheological behavior is also of interest, and has been studied too.

Low molecular weight polybutadienes are used as model materials for better processing and

analytical evaluation.

IV. Introduction of ions through alkylation

Alkylation of amine groups towards ammonium ions

III. Functionalization of intermediate with nucleophile

Epoxide opening by amines as example for functionalization

with nucleophilesRealization of significant conversion

II. Rheological evaluation of intermediate

Estimation of large scale processing feasability

I. Access to a reactive polybutadiene based intermediate

Modification of polybutadiene double bonds by epoxidation Developement of scaleable process

Results and Discussion

24

5 Results and Discussion

5.1 Scalable Semi-Batch Process for Epoxidation of Polybutadienes

Various commercially available polybutadienes with relatively low molecular weights were

used in this study. They have a similar microstructure with a vinyl content below 20% and

almost no cyclic groups (Table 3; Figure 11). They appear as low to medium viscous oils.

The PBus were dissolved in toluene to a weight percentage of 10-15%, being in the range of

concentration in a typical PBu production facility.[5] The resulting viscosity of the solutions

guarantees a sufficient fast mass transport between the phases (vide infra). The phase

relationships of the liquid-liquid system were kept in the range of 5-35 for Vorg/Vaq.

Table 3. Low molecular weight polybutadienes

PBu molecular mass (Mn)

[g/mol]

1,4-DB

[%]

1,2-DB

[%]

cyclic-DB

[%]

cis:trans

ratio2

Lithene PM4 1500 80 19 <1 40:60

Lithene N4-5000 5000 88 12 0 40:60

LBR-307 8000 86 14 0 40:60

LBR-305 26000 91 9 0 40:60

The progress of the epoxidation was analyzed offline from samples taken during the reaction.

Proton NMR spectroscopy is useful to analyze the mixture with respect to conversion of the

double bonds. The 1H-NMR was used as the main method to determine the degree of

functionalization because of its quick and easy operability and its congruence with results

from elemental analyses (vide infra). A typical spectrum with the assignments is shown in

Figure 11. The stacked spectra of the samples show the course of an epoxidation with

samples taken hourly from the reaction mixture.

2 Due to overlapping peaks, the ratios are just approximate values


25

Figure 11. 1H-NMR spectrum of epoxidized polybutadiene (Lithene PM4)

The epoxide content of the polymer was also determined by titration method of JUNG and

KLEEBERG[78] (Figure 12). The titration measurements are only until 15% of conversion in

good accordance with the 1H-NMR based data.

2.6

7

2.9

1

4.9

7

5.3

8

5.4

1

5.5

6

2.53.03.54.04.55.05.56.0f1 (ppm)


26

Figure 12. Titration of epoxidized polybutadiene (Lithene PM4) samples with different degrees of conversion

MALDI-Tof spectra show the expected mass separation of 54 (butadiene; Figure 13) and 70

(butadiene oxide; Figure 14) between the individual signals. The molecular weight of

Lithene PM4 calculated from the distribution in the MALDI-Tof spectrum is 1458 g/mol,

close to the producers provided value of 1500 g/mol. The spectrum of an epoxidized sample

(43.4% by NMR) has many more signals and the intensity of the individual masses is much

lower as expected from a random epoxidation of the double bonds. The differences of 16 Da

in mass corresponding to the consecutive addition of oxygen atoms between signals are

nevertheless recognizable and the signals can be assigned to chains with different degrees of

epoxidation around 40% to 50%. Signals of higher mass seem of lower intensity, possibly

because of instability of the ionized species.[79]


27

Figure 13. MALDI-TOF Analysis of Lithene PM4

Figure 14. MALDI-TOF analysis of an epoxidized polybutadiene sample

The kinetics of the epoxidation of unsaturated rubbers with peracids has been reported upon

for batch reactions.[80–82,3,83] The formation of performic acid is much slower than the

epoxidation of the double bonds, and thus the concentration of formic acid (FA) and hence

performic acid (PFA) will have a quasi steady state. The phase exchange process of

performic acid is presumed fast under the conditions. Reactions at various stirrer frequencies


28

above 300 rpm indeed showed no difference in conversion of the double bonds. It is thus not

necessary to consider the rate of phase transfer. The actual concentration of FA in the

aqueous phase however is of high relevance, and the distribution of FA over the organic and

aqueous phase is complex and needs consideration[84–86] The major reactions are (I) the

general, acid catalyzed, formation of PFA from FA and hydrogen peroxide (HP) in the

aqueous phase (and probably partially in the organic phase), (II) the hydrolysis of PFA to

FA and HP, and (III) the epoxidation of olefins by PFA (Scheme 2).[87]

+ → + − = ∗ ∗ ∗

+ → + − =

+ → + =

↔ , ,

3 → 2 + + + − = Scheme 2. Reactions in the system and their rate laws

The initial rates of epoxidation at high olefin concentration may scale linearly with the rate

of PFA formation. The overall epoxidation is thus initially determined by the rate law of

= ∗ ∗ ∗ ; the concentration of PFA can be considered

constant because of the assumed large differences in the rates of formation and consumption

by hydrolysis and epoxidation. Overall a complex system arises, and it occurred non-trivial

to simulate the process. In general, it seems necessary in simulations to consider many

reactions of the system, also for a course description. The relevant concentration of FAaq and

consequently the pH-value is a function of the phase equilibrium constant and the volume of

the organic and aqueous phase. The phase equilibrium constant may change with conversion

(Scheme 2). In addition, HP is partly soluble in toluene, and may form PFA also in the

organic phase. The decomposition of PFA proceeds along several pathways to several

products, including water, carbon dioxide and oxygen and cannot apriori be neglected

(Scheme 2).[87] The decomposition leads to a decrease in acid concentration and comprises


29

a loss of oxidizing equivalents. Both effects may be relevant to the rate of reaction and

conversion, and become more apparent at higher reaction temperatures, longer reaction times

and higher conversion, in particular at high acid and hydrogen peroxide concentrations. Also

dependencies of the epoxidation rate on olefin conversion have been observed.[3]

5.1.1 Epoxidation with Hydrogen Peroxide Feed

The semi-batch epoxidation of PBu was investigated with a feed of hydrogen peroxide,

maintaining the initial amount of FA. The concentration of FA is therefore consequently

decreasing with reaction time. This is advantageous for preventing side reactions, in

particular for precluding the addition of FA to the epoxidized double bonds, which is also

an acid catalyzed reaction. Thus, the acid concentration is decreasing by conversion and by

dilution as the concentration of the epoxide entities increases. It would consequently allow

a higher initial amount of FA at the start of the PBu functionalization and thus to have higher

initial rates. The conversion of double bonds during an epoxidation reaction in semi-batch

mode with respect to hydrogen peroxide shows an about linear dependency with time up to

a conversion of 15% of the double bonds under the conditions used by DILCHER[88] (Figure

15; Table 4). Performing the reaction in semi-batch mode thus gives a high predictability up

to that level of epoxidation. FA related ring opening reactions leading to hydroxyl carbonyl

entities could not be detected. The initial rate increases linearly with the HP feeding rate to

a rate of 7.23 mL/h. Higher H2O2 (30% in H2O) feeding rates lead to a less than linear

increase. This indicates that the conversion of HP is slower than the feeding rate or a

saturation in PFA sets in. The initial about linear rate of epoxidation continuously decreases

at conversions above 10-15 mol% (Figure 16), in particular noticeable for higher feeding

rates. Course simulations show that this is majorly explainable by the decrease in [FAaq] and

concomitantly the [H+], as has been observed by different HP concentrations in batch

reactions.[3] The concentration of HP increases steadily as the rate of conversion to epoxides

is lower than the feeding rate.


30

Figure 15. Double bond conversion and reaction rates in a semi-batch process (56.6 g Lithene PM4) with various feeds of hydrogen peroxide at 65 °C with 2.03 mL FA in 369 mL toluene (Table 6)[88]

Table 4. Conversion and reaction rates of semi-batch-epoxidations – conditions: 2.03 mL FA, 369 mL toluene at 65 °C[88]

# time [h] final conversion [%] H2O2 feeding rate [ml/h] reaction rate

[mol/L*s]

1 4 14.7 10.19 3.13·10-5

2 5 15.6 7.23 2.64·10-5

3 7 17.1 5.00 1.91·10-5

4 8.5 16.0 2.39 1.35·10-5


31

Figure 16. Double bond conversion of a typical epoxidation of 35.6 g Lithene PM4 in a semi-batch process - conditions: 10 mL/h HP at 64 °C, 3 mL FA and 255 mL toluene

The initial rates were consequently determined in dependence of the FA amounts at a feeding

rate of 0.0981 mol/h HP (Figure 17, Table 5). The rate again shows an initial linear course

followed by a continuously decrease in rate. The initial rates scale in a complex manner with

the amount of FA as expected for the system of reactions in Scheme 2.

Figure 17: Double bond conversion for different amounts of formic acid and 14 wt% of Lithene PM4 in

toluene with 10 mL/h HP at 65 °C (Table 6 entry 33-50)


32

Table 5. Conversion and reaction rates of epoxidation as a function of the FA concentration.

# FA [mol/L] Conversion (6 h) [%] initial reaction rate [mol/L*s]

1 0.11 13.4 3.25·10-5

2 0.22 29.8 5.35·10-5

3 0.44 46.4 6.46·10-5

Table 6: Semi-batch epoxidations of Lithene PM4 in a 1 L steel and glass and 10 L steel reactor at 65 °C

# PBu [g] toluene

[mL]

FA [mL] feeding rate

H2O2 [mL/h]

time [h] conversion

[%]

15 56.6 369 2.03 2.39 1 1.7

16 56.6 369 2.03 2.39 2.5 4.8

17 56.6 369 2.03 2.39 4 8.0

18 56.6 369 2.03 2.39 5.5 10.6

19 56.6 369 2.03 2.39 7 13.3

20 56.6 369 2.03 2.39 8.5 16.0

21 56.6 369 2.03 5.00 2 5.0

22 56.6 369 2.03 5.00 4 10.2

23 56.6 369 2.03 5.00 6 14.9

24 56.6 369 2.03 5.00 7 17.1

25 56.6 369 2.03 7.23 1 3.1

26 56.6 369 2.03 7.23 2 6.7

27 56.6 369 2.03 7.23 3 10.1

28 56.6 369 2.03 7.23 5 15.6

29 56.6 369 2.03 10.19 1 3.2

30 56.6 369 2.03 10.19 2 7.6

31 56.6 369 2.03 10.19 3 11.5

32 56.6 369 2.03 10.19 4 14.7

33 35.6 255 3.00 10.0 1 7.2

34 35.6 255 3.00 10.0 2 14.7

35 35.6 255 3.00 10.0 3 19.5

36 35.6 255 3.00 10.0 4 24.1

37 35.6 255 3.00 10.0 5 27.1


33

# PBu [g] toluene

[mL]

FA [mL] feeding rate

H2O2 [mL/h]

time [h] conversion

[%]

38 35.6 255 3.00 10.0 6 29.8

39 37.1 263 1.55 10.0 1 2.9

40 37.1 263 1.55 10.0 2 6.2

41 37.1 263 1.55 10.0 3 8.6

42 37.1 263 1.55 10.0 4 10.3

43 37.1 263 1.55 10.0 5 11.8

44 37.1 263 1.55 10.0 6 13.4

45 35.5 255 6.00 10.0 1 9.8

46 35.5 255 6.00 10.0 2 19.4

47 35.5 255 6.00 10.0 3 29.8

48 35.5 255 6.00 10.0 4 36.5

49 35.5 255 6.00 10.0 5 41.4

50 35.5 255 6.00 10.0 6 46.4

51 679 4480 48.8 50.0 1.67 3.2

52 679 4480 48.8 50.0 3 6.2

53 679 4480 48.8 50.0 4.5 9.2

54 679 4480 48.8 50.0 6 12.1

55 679 4480 48.8 50.0 7.5 15.0

56 679 4480 48.8 60.0 1.5 3.8

57 679 4480 48.8 60.0 3 8.3

58 679 4480 48.8 60.0 4.5 11.8

59 679 4480 48.8 60.0 6 15.0

60 679 4480 48.8 60.0 7.5 18.1

The temperature dependence of the rate of epoxidation of Lithene PM4 was determined

between 54 and 80 °C at higher temperatures than reported upon in the literature [2,44,50,3,83]

The initial concentration of FA was 0.27 mol/L and the rate of feeding HP was 18 mol%/h

in relation to 1,4-double bonds. The conversion of PBu after 6 h increases with temperatures

up to 75°C (Figure 18). The observed rate constants as function of the temperature are

indicating an overall activation energy for the process of about 29 kJ/mol. The somewhat


34

lower activation energy compared to published numbers might be attributed to the lower

molecular weight of the polybutadienes used in this study.[3],[3] Higher temperatures lead to

a lower conversion. Evidence for epoxide opening reactions could not be found in any of the 1H-NMR spectra of the samples. It has been reported that the decomposition rate of PFA has

a stronger temperature dependence than the in-situ formation of the same[89]. Thus the

decrease in the degree of epoxidation could be related to the increasing rate of decomposition

of PFA or H2O2, due to a resulting decrease of PFA and FA concentration in the system

(Scheme 2).[87] The progress of conversion was monitored at four of the eight reactions

(Figure 18) of which the maximum rate constants are listed in Table 7.

Figure 18. Temperature dependency of epoxidation degree with Arrhenius-plot – conditions: 16 wt% Lithene PM4 in toluene and 1.5 wt% FA, 18 mol% HP in relation to 1,4-double bonds per hour (Table 8)

Table 7. Rate constants and conversions of epoxidations at various temperatures

temperature [°C] final conversion [%] rate constant k [L/mol*s]

54 21.2 9.67·10-5

64 29.8 1.38·10-4

75 45.9 1.83·10-4

80 36.7 1.54·10-4


35

Table 8. Semi-batch epoxidations of Lithene PM4 in a 1 L glass reactor at different temperatures

# PBu [g] Temper-

ature [°C]

toluene [mL] FA [mL] feeding rate

H2O2 [mL/h]

time [h] conver-

sion [%]

61 35.4 54 256 3 10 1 3.7

62 35.4 54 256 3 10 2 8.8

63 35.4 54 256 3 10 3 12.4

64 35.4 54 256 3 10 4 15.8

65 35.4 54 256 3 10 5 18.3

66 35.4 54 256 3 10 6 21.2

67 35.6 64 255 3 10 1 7.2

68 35.6 64 255 3 10 2 14.7

69 35.6 64 255 3 10 3 19.5

70 35.6 64 255 3 10 4 24.1

71 35.6 64 255 3 10 5 27.1

72 35.6 64 255 3 10 6 29.8

73 64.6 70 475 6 18 6 37.3

74 65.2 73 474 6 18 6 42.0

75 61.6 75 450 5.5 17 1 11.4

76 61.6 75 450 5.5 17 2 21.9

77 61.6 75 450 5.5 17 3 30.1

78 61.6 75 450 5.5 17 4 37.3

79 61.6 75 450 5.5 17 5 42.1

80 61.6 75 450 5.5 17 6 45.9

81 64.8 77 495 6 18 6 40.3

82 65.3 80 477 6 18 1 7.1

83 65.3 80 477 6 18 2 18.5

84 65.3 80 477 6 18 3 25.5

85 65.3 80 477 6 18 4 30.6

86 65.3 80 477 6 18 5 34.0

87 65.3 80 477 6 18 6 36.7

88 64.1 80 484 6 18 6 35.8


36

5.1.2 Larger Scale Reactions

The reaction conditions of the semi-batch operation in form of polybutadiene concentration,

formic acid concentration and the relative feeding rate of hydrogen peroxide were transferred

to larger scale. A temperature of 65 °C was chosen for the scale up reactions, somewhat

below the maximum rate at 75 °C where the decomposition of PFA becomes noticeable. It

was found, that reactions in a 10 L reactor are very comparable to those in a 1 L steel reactor

(Figure 19).[88] Again, a linear increase of double bond conversion was found with time and

feeding rate. The observed rate constants are basically the same even though a higher

concentration in FA is needed. The process is thus easily scaled up.

Figure 19. Double bond conversion of 56.6 g and 679 g Lithene PM4 in a 1 L (top) and 10 L (bottom) steel reactor respectively with various dosing rates - conditions: 65 °C and 2.03 mL and 48.8 mL in 369 mL

and 4482 mL toluene respectively[88]


37

5.1.3 Higher molecular weight polybutadienes

The process described was also applied to the other polybutadiene samples (Table 3). The

differences in microstructure and molecular mass of the PBus of this study are of minor

importance for the rate of conversion in the range considered here (Figure 20; Table 9). The

higher viscosity of the toluene solution of PBu with a higher molecular mass is not

influencing the reaction rate. The formation of PFA remains the rate determining step of the

reactions in Scheme 2.[87] Likewise, the cis/trans ratio of the 1,4-double bonds is not

substantially changed by the epoxidation. Former studies revealed a rate of epoxidation

descending in the order cis-1,4 > trans-1,4 > vinyl-1,2.[2] Indeed the cis-1,4 entities are

somewhat faster epoxidized than trans-1,4 units (Table 10). None of the polybutadienes in

this work showed an epoxidation of vinyl groups within the degrees of epoxidation reached.

The vinyl double bond content of maximum 19% might be too low to detect epoxidation,

which will be negligibly.

Figure 20. Double bond conversion of different polybutadienes in a 10 L steel reactor - conditions: 75 °C, 18 wt% PBu in toluene and 1.5 wt% FA, 0.3 mol% HP in relation to double bonds per hour (Table 11)

Table 9. Reaction rates of different molecular weight polybutadienes.

molecular mass [g/mol] conversion (7h) [%] initial reaction rate [L/mol*s]

1500 43.4 4.73·10-5

5000 40.9 4.79·10-5

8000 40.6 4.77·10-5

26000 41.1 4.75·10-5


38

Table 10. Cis:trans selectivity of 1,4-epoxidation

PBu degree of epoxidation [%] cis:trans ratio

Lithene PM4 22.5 48:52



Lithene N4-5000 21.2 51:49

Lithene N4-5000 32.1 50:50

Lithene N4-5000 40.9 49:51

LBR-307 20.9 50:50

LBR-307 32.5 49:51

LBR-307 40.6 48:52

LBR-305 20.6 51:49

LBR-305 32.4 50:50

LBR-305 41.1 49:51

Table 11. Semi-batch epoxidations of different polybutadienes in a 10 L steel reactor at 75 °C

# PBu PBu [g] toluene

[mL]

FA

[mL]

feeding rate

H2O2 [mL/h]

time [h] conversion

[%]

89 PM4 1002 6514 82 200 3 22.5

90 PM4 1002 6514 82 200 5 33.8

91 PM4 1002 6514 82 200 7 43.4

92 N4 1000 6514 82 200 3 21.2

93 N4 1000 6514 82 200 5 32.1

94 N4 1000 6514 82 200 7 40.9

95 LBR-307 700 4561 58.2 140 3 20.9

96 LBR-307 700 4561 58.2 140 5 32.5

97 LBR-307 700 4561 58.2 140 7 40.6

98 LBR-305 700 4562 57.4 140 3 20.6

99 LBR-305 700 4562 57.4 140 5 32.4

100 LBR-305 700 4562 57.4 140 7 41.1


39

5.1.4 Material Properties

The introduction of epoxide groups increases the rigidity and polarity of the polymer, and

therefore is increasing the glass transition temperature (Figure 21).[3] The glass transition

temperatures of the various PBus are about the same, and this holds also for the epoxidized

derivatives as along as the degree of epoxidation is comparable (Table 12).[3,90] The minor

differences are related to differences in the microstructure, i.e. the content of 1,2-vinyl

entities (Table 3). The increase of the glass transition temperature (Tg) follows a general Fox

equation (eq. 1, Figure 21). An increase in Tg of 0.83-0.93 K per mol% of epoxide groups is

found, in accordance with prior results[3].

,100% ,0%

1 epoxides polybutadiene

g g g

w w

T T T (1)

wepoxides and wpolybutadiene define the ratio of weight of epoxidized and non-epoxidized

butadiene building blocks, Tg,100% and Tg,0% represent the Tg of PBu with all double bonds

oxidized and non-epoxidized PBu respectively.

Table 12. Glass transition temperatures of investigated polybutadienes

PBu molecular mass (Mn) [g/mol] glass transition (Tg) [K]

Lithene PM4 1500 181.9

Lithene N4-5000 5000 181.3

LBR-307 8000 180.6

LBR-305 26000 178.7


40

Figure 21. Increase in Tg with degree of epoxidation following the fox-equation

Table 13. Increase of Tg per percentage point of epoxidation

molecular weight (Mn) [g/mol] increase in Tg per mol% of double

bond epoxidation [K]

1500 0.83

5000 0.88

8000 0.91

26000 0.93

The bulk viscosity of the polybutadienes increases about one order of magnitude by

epoxidizing to about 50 mol% (Figure 22; Table 15).[16] This is again related to change in

chain stiffness and the concomitant increase in glass temperature.[91]


41

Table 14. Melt viscosities of polybutadiene of this study at 25°C.

PBu molecular weight (Mn) [g/mol] viscosity [Pa*s]

Lithene PM4 1500 0.7-0.95 3

Lithene N4-5000 5000 3-5 3

LBR-307 8000 1.5 4

LBR-305 26000 40 4

Figure 22. Viscosity [η*] at 1.6 Hz with increasing degree of epoxidation of Lithene PM4 at room temperature.

Table 15. Complex oscillatory viscosity of PBu with various degrees of epoxidation

conversion [%] complex oscillatory viscosity [Pa*s]

0 0.96

7.6 1.48

13.4 1.83

21.2 3.18

29.8 4.23

37.3 6.83

45.9 13.79

3 Melt viscosity at 25 °C 4 Melt viscosity at 38 °C


42

5.1.5 Summary

The epoxidation of low molecular weight polybutadienes using a semi-batch process was

reported in this work. The temperature of 75 °C is particular useful for the reaction.

Formation of performic acid is rate determining at this temperature, and the decomposition

rate is low and the ring opening reaction of the product is negligible. The process was

successfully scaled up to a reaction volume of 10 L obtaining 1 kg of product material. The

reaction of various molecular weights between 1500 and 26000 g/mol showed a universal

applicability of the process in this range. The increase in glass transition temperature follows

the Fox equation and the viscosity increases roughly by one order of magnitude to about

50% of epoxidation.


43

5.2 Rheological Properties of Epoxidized Liquid Rubber in Bulk and Solution

It is essential for material processing to know about the material properties. The knowledge

of the viscosity of a material can be vital for processing steps such as mixing or pumping.

Therefore the prediction of the viscosity in synthetic steps is of high interest due to the layout

of equipment. In this work a model for the prediction of the complex oscillatory viscosity

(η*) of low molecular weight polybutadienes is developed. This model allows the prediction

of complex oscillatory viscosity as a function of molecular weight, epoxidation ratio,

concentration in toluene and temperature.

Experimental data[92,93] support the approach to factorize the viscosity into a temperature-

and concentration-dependent friction factor ζ(T,Φ) and a molecular weight- and

concentration-dependent structural (topology) factor g(M,Φ). The viscosity can therefore be

expressed as:

= , ∙ , (2)

The viscosity was determined using rheology with oscillatory shear. A typical measurement

is depicted in Figure 23a. The measurement shows Maxwell type behavior as expected for a

viscoelastic fluid. The viscosity is independent of the applied frequency (Figure 23),

showing that the PBus can be considered newtonian liquids.

The frequency region relevant for extrusion, mixing or pipe flow is 4 -110 10 [rad s ] [77].

The viscosity of the investigated polybutadienes in this region is 1 430 Pa s (Figure 23b).


44

Figure 23. a) Oscillatory shear experiment of LBR-300 (MW = 44000) at 45 °C, b) Complex viscosities of polybutadienes with different molecular masses

5.2.1 Molecular Weight Dependance

The viscosity in the region depicted in Figure 23 is independent of frequency and therefore

equates the zero-shear-viscosity (η0) which is proportional to the weight average molecular

mass (Figure 24) in the usual power law of

∝ (3)

below and

∝ . (4)

above the entanglement molecular mass.


45

Figure 24. Zero-shear-viscosity in relation to molecular weight in bulk at room temperature

The expected slope of viscosity versus molecular mass is 1.0 below the entanglement

molecular mass and 3.4 above it. RENDELL and NGAI found the critical molecular mass at

6400 g/mol for their system[94] and indeed the slope in Figure 24 is slightly higher than 1.0

even for the part of lower molecular weight. However, since the molecular masses

investigated are still very low and the influence of entanglements is expected to be low in

this case, an average slope of 1.5 was chosen for the following discussion. This would

adequately describe the transition region.

5.2.2 Concentration Dependance

The polybutadiene samples were measured in bulk as well as volume fractions of 0.75 and

0.5 in toluene which is a common solvent for polybutadiene synthesis.[5] The slope of the

viscosity vs. molecular weight plot is independent of the volume fraction in toluene but is

shifted towards lower viscosities at lower concentrations as expected for these molecular

masses. At higher molecular masses the slope would also be expected to be concentration

dependent[95]. The viscosity shift covers roughly one magnitude per 0.25 of toluene volume

fraction (Figure 25a).


46

Figure 25. a) Viscosity developement with molecular weight at different volume fractions in toluene b) Mastercurve of the shifted viscosities

If a shift factor of φ3.4 is considered, these three slopes can be transformed on to a master

curve (Figure 25b). The proportionality of the zero shear viscosity could ergo be described

as follows:

∝ . ∙ . (5)

Knowing the concentration (volume fraction, Φ) and molecular weight of a polybutadiene

sample, it is possible to predict the viscosity of the solution of unfunctionalized

polybutadiene in this molecular weight region with this master curve.

5.2.3 Functionality dependence

The epoxidation ratio needs also be considered if the polybutadienes are functionalized. The

epoxidation ratio is influencing the viscosity of the polybutadienes due to changes in chain

stiffness and polarity (s. 5.1.4). The degrees of freedom are reduced with an epoxide group

compared to the double bond. Additionally, the more polar epoxide groups are increasing

the chain-chain interaction parameter.[96] The viscosity is therefore gradually increasing

during epoxidation reactions. To take these changes into account a shift factor for the

changes in viscosity due to the epoxidation ratio (f) was introduced:

∝ + . ∙ . (6)


47

Figure 26. a) Viscosity developement with molecular weight at different epoxidation ratios b) Mastercurve of the shifted viscosities

The slope again is the same for all functionalization ratios but the curves are shifted with

(1+f)3.4.

5.2.4 Interim Conclusion

The dependence of the viscosity on molecular weight, concentration and epoxide ratio can

be merged into one equation:

∝ + . ∙ . ∙ . (7)


48

Figure 27. Mastercurve considering degree of epoxidation, concentration and molecular weight

The influence on viscosity by the epoxidation ratio is based on the changes in structure whilst

the influence by the concentration in toluene is based on the solution states of polymer

solutions.

The viscosity can therefore be shifted onto a master curve taking the epoxidation ratio and

concentration into account. Furthermore, the viscosity during epoxidation is now predictable

with the knowledge of concentration and expected epoxidation ratio.

Additionally, the temperature dependence is needed for the prediction of viscosities at

reaction temperatures other than room temperature.

5.2.5 Temperature Dependence

The glass transition temperatures for the polybutadiene samples in this work are below 0 °C

and therefore low enough for the ARRHENIUS law to be applicable.

The temperature dependence of the zero shear viscosity shows a similar progress for all

samples (Figure 28).


49

Figure 28. a) Temperature dependence of viscosity at 26000 g/mol, b) Temperature dependence of viscosity at different molecular weights

Following the ARRHENIUS law, a shift factor (aT) for the temperature dependence can be

introduced:

a 0T T

0 0 0

( )1 1exp with

( )

E Ta a

R T T T

(8)

The reference temperature T0 was set to 25 °C and the resulting ARRHENIUS plot shows no

dependence of the flow activation energy (Ea) on molecular weight of the polybutadienes

(Figure 29).

Figure 29. ARRHENIUS plot of the tested polybutadiene samples


50

However, the epoxidation ratio has an influence on the flow activation energy (Figure 30).

Figure 30. Functionalization dependence of the flow activation energy

These dependencies were expected since the flow activation energy is a function of chain

rigidity and degree of freedom and the epoxide groups increase the chain stiffness. The flow

activation energy also is concentration dependent, which leads to the following relation:

= . − − ∙ . + . ∙ (9)

Uniting all influencing parameters into one equation, it is now possible to predict the

viscosity of a polybutadiene system with the knowledge of epoxidation ratio, concentration,

molecular weight and temperature of the system:

∝ . ∙ . ∙ . (10)

The flow activation energy (Ea) is a function of the concentration (Φ) as well as the

epoxidation ratio (f).

5.2.6 Conclusion

The present model can be used to predict the viscosity evolution during epoxidation reactions

of low molecular weight polybutadienes. This equation enables the operator to design the

optimal reactor system for the processing of the reaction mixture during epoxidation of

10 30 500 20 400

10

20

30

40

50

60

Ea

[kJ*m

ol-1

]

f [%]

0.5 0.6 0.7 0.8 0.9 1.00

10

20

30

40

50

60

Ea

[kJ/m

ol]


51

polybutadienes. The dependencies of the zero shear viscosity on the epoxidation ratio as well

as the concentration are to the power of 3.4 whilst the molecular weight dependence is to the

power of 1.5 most likely due to the lack of entanglements at lower molecular weights.

The temperature has its influence through the ARRHENIUS law and the flow activation

energy. This relation is shown in the overall equation for the prediction of the shear viscosity

of low molecular weight polybutadienes

∝ . ∙ . ∙ .. ∙ . . ∙ (11)


52

5.3 Aminolysis of Epoxidized Polybutadienes Catalyzed by Lithium Bromide

The synthesized epoxidated polybutadienes were opened by nucleophilic addition of amines.

Butyl amine was chosen for the addition because of its ability to dissolve polybutadiene,

good NMR signature and relatively high nucleophilicity. Lithium bromide was chosen as

Lewis acid catalyst because it is inexpensive, nontoxic, oxophilic and has good availability.

The reactivity of different epoxide isomers in this system was first evaluated with a model

system.

5.3.1 Reaction Mechanism

Lithium bromide is an inexpensive, easy to handle, non-toxic Lewis acid. The additionally

high affinity of lithium to oxygen makes lithium bromide an excellent choice catalyst for the

aminolysis of epoxides.[56]

Scheme 3. Lithium bromide catalyzed aminolysis

Lithium bromide coordinates with the oxygen atom of the epoxide group forming a donor-

acceptor complex. This complex is more susceptible to nucleophilic attack because of the

carbon oxygen bond weakening. The rate determining step most likely is the attachment of

the amine.[97]

5.3.2 Analysis of a Model System

The aminolysis was initially carried out using a set of simple epoxides to better understand

the relative reactivities of asymmetric and symmetric cis and trans epoxides. The reaction

was carried out using 1.0 mL (11.2 mmol) of butylene oxide mixture, 1.0 eq. of n-

butylamine in relation to epoxide groups and 10 mol% of lithium bromide in relation to

epoxide groups as catalyst.

The mix of epoxides consisted of 1,2-butylene oxide, cis-2,3-butylene oxide and trans-2,3-

butylene oxide with a ratio of 1:3:6. The 1H-NMR spectrum shows the good distinguishable

signals of the different epoxides (Figure 31).


53

Figure 31. 1H-NMR spectrum of the mix of different butylene oxides

Adding butylamine as well as LiBr to the mixture at room temperature and recording another 1H-NMR spectrum after 30 min, already shows an ongoing reaction of butylamine with the

asymmetric 1,2-butylene oxide (Figure 32).

Figure 32. 1H-NMR spectrum of the butylene oxide mixture with butyl amine added at room temperature

The spectrum shows a conversion of approximately 50% of the 1,2-butylene oxide whereas

the 2,3-butylene oxides show conversions below 1%. Additionally roughly 90% of the 1,2-

butylene oxide opening leaves the amine substituent at the less hindered 1 position.


54

After 150 h at room temperature another spectrum shows a complete conversion of the 1,2-

butylene oxide as well as roughly 70% and 35% conversion of the cis- and trans-butylene

oxide respectively. Taking the different concentrations into account, the reaction rates can

be sorted to the following order: 1,2-butylene oxide >> cis-2,3-butylene oxide > trans-2,3-

butylene oxide.

The results can only partly be transferred to the epoxidized polybutadiene, since the

differences in structure of the polymer chain also influence the reactivity of the different

isomers.[98]

The reactivity of epoxide groups on the polybutadiene backbone was therefore low for the

reaction conditions of the model system to be applied here. The butyl amine was used in

high excess (usually around 5-10 eq in relation to double bonds). Additionally, temperatures

of 90-150 °C were applied. A glass autoclave was used for these reactions.

5.3.3 Progression of Aminolysis with Butyl Amine

Several reactions with different reaction times were conducted to follow the course of the

reaction, given by the difficulties of sample taking in the high pressure process. The

aminolysis was conducted at 90 and 150 °C with different concentrations of catalyst and

with various polybutadienes epoxidized to a different extent.


55

Figure 33. Yield-time curves of Lithene PM4 (with 30% epoxide functionalization) amination with butyl

amine at 150 °C

Figure 34. Yield-time curves of Lithene PM4 (with 42% epoxide functionalization) amination with butyl amine at 90 °C


56

Table 16. Conversion of epoxide groups measured by different methods (30% epoxidized BR, 150 °C)

c(LiBr) [mol%] reaction time [h] conversion

1H-NMR Elemental analysis 13C-NMR IG

- 6 7.1% 7%

- 24 26.0% 23%

- 48 46.6% 40% 47.8%

- 72 64.0% 54%

- 96 70.7% 63%

- 120 74.0% 67%

30 1.5 53.7%

30 3 61.1%

30 6 70.1% 65%

30 24 78.5% 75%

30 48 78.8% 73%

30 72 78.5% 75%

Table 17. Conversion of epoxide groups (42% epoxidized BR, 90 °C)

catalyst [mol%] reaction time [h] conversion (1H-NMR)

- 6 2.7%

- 24 5.6%

- 72 12.1%

4 6 9.2%

4 24 23.6%

4 120 58.4%

30 6 23.5%

30 24 56.9%

30 72 68.5%

30 96 69.6%


57

The results of 1H-NMR spectroscopy and elemental analysis are in good accordance. The

lower conversions calculated by elemental analysis are most likely because of the

hygroscopic nature of the aminated rubbers. A small water content increases the oxygen

content and therefore falsifies the calculated conversion.

The kinetics of aminolysis has been thoroughly discussed in the literature.[99,100] The system

is expected to undergo an SN2 like mechanism. However, a reaction order of two is only

observed under conditions with very low concentrations of hydroxyl-containing entities.

Usually the observed reaction order is between one and two with respect to the amine

resulting of two competing mechanisms:[99]

The reaction is first order with respect to the amine if hydroxyl-containing impurities

are present because of their promotion of the nucleophilic attack (trimolecular

mechanism, Scheme 4a).

The amine proton is acting as an electrophile promoting the attack on the epoxide

group in addition from its role as a nucleophile if no hydroxyl-containing impurities

are present (bimolecular mechanism, Scheme 4b).

Scheme 4. a) Trimolecular mechanism with hydroxyl impurities present, b) bimolecular mechanism without hydroxyl impurities present.[99]

The reaction order is also temperature dependent, as the activation energies of the above

mechanisms are different (depicted activation energies are for the aminolysis of

phenylglycidyl ether). Higher temperatures are therefore reducing the reaction order with

respect to the amine.[99,100] At temperatures used in this work the reaction order in respect to

the amine should be one. In this work the amine is used in high excess and its concentration

is therefore considered constant during the reaction. The reaction should ergo follow a

pseudo zero order kinetics with respect to the amine and only be limited by the epoxide

concentration.


58

The lithium cation is expected to have a higher affinity to the epoxide oxygen relative to the

amine or hydroxide impurities. Therefore the catalyzed reaction pathway should have a

lower activation energy leading to a second order reaction, or with an excess of amine, a

pseudo first order reaction.

The order of the aminolysis is affected by a competing cyclization reaction (see 5.3.4). The

ratio of cyclization is expected to increase with the progress of the reaction since one of the

neighboring epoxide groups needs to be opened prior to a possible cyclization.

The overall order of the reaction is therefore expected to be of pseudo first order at low

conversions changing to a more complex mixed order at higher conversions when the effect

of cyclization increases. That is why an evaluation of the rate constants was only carried out

at the start of the reaction until roughly 40% conversion.

Table 18. Initial rate constants of the aminolysis with butyl amine

cat. [%] Temp. [°C] k [1/s]

0 90 4.7E-75

4 90 3.0E-65

30 90 9.2E-65

0 150 3.6E-66

30 150 1.4E-46

Lithium bromide strongly increases the reaction rate. At 150 °C the reaction with 30 mol%

lithium bromide is almost 40 times faster than without it. At 90 °C 30 mol% of lithium

bromide still inceases the rate constant by a factor of 20.

5.3.4 Intramolecular Cyclization

The yield-time curves of the epoxide opening with butyl amine are heading towards a

maximum below full conversion. 13C-NMR I-Gated measurements however show no

remaining epoxide groups. Therefore, a side reaction will have occured in which the epoxide

groups are opened without additional consumption of butyl amine.

GELLING and others discovered an internal cyclization of highly epoxidated natural rubber

during epoxide opening with hydroxyl functionalities.[101–103] In this cyclization a hydroxyl

5 42% epoxidized polybutadiene 6 30% epoxidized polybutadiene


59

functionality, deriving from epoxide opening, attacks a neighboring epoxide entity forming

a five-membered cyclic ether.

Transferring this reactivity to the present system, two cyclization reactions seem plausible

(Scheme 5).

Scheme 5. Internal rearrangement of epoxide groups

Which cyclization reaction is occurring depends on the position the amine has added to in

the previous epoxide opening and therefore which nucleophilic group is adjacent to the

neighboring epoxide. Since no real differences between the epoxides carbon atoms exist, a

statistical distribution seems likely. Looking into the 13C-NMR I-Gated spectra the signals

between 69 and 72 ppm are attributed to the oxygen neighboring carbon atoms of the THF

ring[104,105] whereas the nitrogen neighboring carbons would be expected somewhat further

upfield where at 62 ppm a corresponding peak is observed (Figure 35). The 1H-NMR

spectrum doesn’t give much information on account of overlapping signals, but resonances

could be found at the expected positions (Figure 36). The expected statistical cyclization is

backed up from the observation of the integrals of the 13C-NMR reverse I-Gated spectrum.


60

Figure 35. 13C I-Gated NMR spectrum of aminated polybutadiene rubber

Figure 36. 1H-NMR spectrum of aminated polybutadiene rubber (detail)

Since the epoxidation of polybutadiene under the used conditions is statistically, the

probability of neighboring epoxide groups, allowing such cyclization, increases with

increasing epoxidation ratio. A higher relative butyl amine functionalization could indeed

be observed with lower epoxidated polybutadienes (Figure 37).


61

Figure 37. Degree of aminolysis by amine content with Lithene PM4 with different degrees of epoxidation

In conclusion the remaining epoxide groups not getting opened by the primary butyl amine

are most likely undergo an internal cyclization where either the secondary amine or the

hydroxyl entity attached to the polymer chain is attacking a neighboring epoxide group.

The linear fit of Figure 37 crosses the Y-axis at 100% as expected since without epoxide

groups no neighboring epoxides for cyclization exist. At 100% epoxidation there should be

a 100% chance of two epoxide groups to have a neighboring epoxide group, which would

lead to 100% heterocycles. This would only be true for 100% 1,4-polybutadiene without any

branching or 1,2-vinyl groups. The theoretical heterocyclic content of 75% at 100%

epoxidation therefore seems a reasonable number.

The ratio of cyclization is not temperature dependent at temperatures above 90 °C. The

aminolysis with a 42% epoxidized sample at 90 °C and 150 °C gave the same maximum

amine introduction of 70%. This fact leads to the conclusion that the activation energy of the

cyclization reaction is lower than the activation energy of the aminolysis with butylamine.

A reason for the lower activation energy could lie in the favorable five membered transition

state of the intramolecular cyclization.

The concentration of lithium bromide also doesn’t influence the maximum amine

introduction. Reactions with the same epoxidized polybutadiene but different LiBr

concentrations gave the same maximum conversion but different reaction rates. This also


62

was expected because the lithium cation is expected to attach to the oxygen atom weakening

the oxygen carbon bond and making it more vulnerable for nucleophilic attacks independent

of the type of attacking nucleophile. Hence the ratio of cyclization is only affected by the

concentration of neighboring epoxide groups.

5.3.5 Higher Molecular Mass Epoxidized Polybutadienes

In the same process polybutadienes with a similar configuration but higher molecular mass

were aminated under the same conditions. The polybutadienes all had a similar epoxidation

grade (about 40%) and were treated with 13 eq of butyl amine and 30 mol% of lithium

bromide at 150 °C for 24 h. The conversion of aminolysis show basically no differences

(Table 19).

Table 19. Aminolysis of epoxidized polybutadienes of different molecular masses

Initial molecular mass (Mn) [g/mol] Epoxide ratio [%] Conversion [%]

1500 41.1 68.6

8000 39.0 73.5

26000 40.5 73.2

The somewhat higher conversion with the polybutadiene of 8000 and 26000 g/mol might

result from differences in the molecular configuration (see Table 20). The lower content of

1,2-double bonds would lower the probability of two neighboring epoxide groups, which

leads to a lower probability of cyclization (see 5.3.4) and ergo a higher possible amine

concentration.

5.3.6 Material Properties

The introduction of amine groups increases the polarity of the polymer increasing chain-

chain interactions and therefore increases the glass transition temperature (Figure 38). The

fitting curve represents the results of the Fox equation, only in accordance with the

experimental results, if a lower glass transition temperature for the epoxidized polybutadiene

starting material is used. A change in polymer-chain-conformation with introduction of

amine groups might be a reason for this deviation.


63

Figure 38. Glass transition temperatures of Lithene PM4 with different degrees of aminolysis

The molecular masses by GPC seem to decrease with increasing amine content (Figure 39).

The GPC elugrams were recorded against a polystyrene standard in THF. With ongoing

reaction the aminated rubber becomes more and more dissimilar to polystyrene and probably

shows different coil expansion behavior. The seeming decrease in molecular mass (Mn) to

almost 200 g/mol would equal 1-4 repeating units in a chain. This drastic decrease in chain

length wouldn’t go unnoticed in 1H-NMR or DSC measurements which look very similar

independent of the degree of aminolysis.

Figure 39. Molecular masses (Mn) by GPC (■) and theoretical progression of the molecular mass (─)


64

The molecular mass was also measured by vapor pressure osmometry in THF against a

standard of benzil (210.23 g/mol). For verification a sample of Lithene PM4 was also

measured. The expected molecular masses were 1500 g/mol (Lithene PM4) and 2300 g/mol

(aminated BR). The values measured by vapor pressure osmometry were 1670±115 g/mol

(Lithene PM4) and 1985±77 g/mol (aminated BR).

MALDI-Tof spectra of the aminated BR show relatively low intensity, which makes it

difficult to calculate a molecular weight distribution. The flight pattern of the chains might

represent a false molecular weight distribution due to different stabilities of the chains. The

high molecular weight signals are difficult to separate from the noise. The best results were

achieved with Dithranol as matrix and sodium trifluoroacetate as added salt (Figure 40).

Nevertheless, the spectrum shows molecular masses and a distribution of the same in the

expected range. The signals at 685 and 2441 m/z respectively belong to the matrix.

Figure 40: MALDI-Tof spectrum of aminated BR

5.3.7 Summary

The aminolysis of epoxidated low molecular weight polybutadienes using a lithium bromide

catalyst was reported in this work. A solvent free process with elevated temperatures was

used to achieve high yields at moderate reaction times. The maximum yield depends on the

epoxidation ratio of the starting material, due to an intramolecular cyclization reaction. The


65

process was successfully applied to molecular masses of epoxidized polybutadiene of up to

26000 g/mol. The change in polarity and rigidity of the chains reduces the hydrodynamic

volume as well as increases the glass transition temperature of the aminated products.


66

5.4 Quaternization of aminated polybutadienes

The aminated polybutadienes were treated with methyl iodide to quaternize the amine

functions, creating quaternary ammonium ions. Methyl iodide has a sterically low

hinderance, and iodide is a good leaving group making methyl iodide an excellent

methylating agent.[106] Potassium carbonate was chosen as base, which can promote the

quaternization of amines at room temperature.[69]

5.4.1 Reaction Mechanism

The amine is attaching to the electrophilic carbon of iodomethane by nucleophilic attack.

The reaction is following SN2 mechanism because of the high nucleophilicity of secondary

amines. In a second step potassium carbonate is binding the amine proton forming potassium

bicarbonate and potassium iodide. The tertiary amine is usually further methylated to the

quaternary ammonium ion with monopotassium carbonate as counterion.[69,107]

Scheme 6. Methylation of secondary amines by iodomethane and potassium carbonate

The aminated polybutadienes also contain hydroxyl functionalities which can be methylated

too. Theoretically the nucleophilicity of alcohols is much lower compared to that of

secondary amines and should therefore react much slower with iodomethane. Nevertheless,

a set of simple aminoalcohols was treated with iodomethane and potassium carbonate to get

a feeling of reactivity under given reaction conditions.

5.4.2 Analysis of a Model System

Two different aminoalcohols were used for this test reaction. 2-(Dimethylamino) ethanol

and 2-amino-2-methylpropan-1-ol were treated with one and three equivalents of

iodomethane respectively in the presence of one and three equivalents potassium carbonate

for 24 h at room temperature in THF. The reaction was quenched with aqueous ammonia


67

solution after 24 h. 2-(Dimethylamino) ethanol was completely quaternized which was

accompanied by a downfield shift of the methyl signals. No sign of alcohol methylation and

therefore ether formation was visible (Figure 41).

2-Amino-2-methylpropan-1-ol was quaternized to 75% under the described conditions. The

remaining 25% stopped at tertiary amine. Evaporation of iodomethane (bp.: 42 °C[108]) might

be a reason for the incomplete quaternization. No sign of ether formation was visible (Figure

42).

Figure 41. 1H-NMR spectrum of quaternized 2-(dimethylamino) ethanol


68

Figure 42. 1H-NMR spectrum of quaternized 2-amino-2-methylpropan-1-ol

Even the sterically more hindered 2-amino-2-methylpropan-1-ol was readily quaternized

under given conditions. The reaction rate was probably lower compared to that of 2-

(Dimethylamino) ethanol, but the alkylation was still highly selective towards amine

methylation and no primary or secondary amine was left unreacted.

These reaction conditions were transferred to the quaternization of aminated polybutadienes.

5.4.3 Progression of Quaternization with Iodomethane

The quaternization of 2-amino-2-methylpropan-1-ol showed an insufficient conversion to

ammonium ions with three equivalents of iodomethane. The methylation of aminated

polybutadienes carrying secondary amines would need at least two equivalents of

iodomethane to give the quaternary ammonium ion. An excess of iodomethane (5-10 eq

relating to the concentration of amine groups) was used for the quaternization of aminated

polybutadienes to avoid insufficient methylation like seen with 2-amino-2-methylpropan-1-

ol. Iodomethane was divided into two halves, to compensate for the probable evaporation of

iodomethane over the course of the reaction. The first half was added at the start of the

reaction, the other half after a certain time of reaction.

The aminated polybutadiene was dissolved in THF and potassium carbonate was dispersed

in the solution under stirring. Iodomethane was added to start the reaction. Samples were


69

taken during the reaction. After 24 to 72 h the dispersion was treated with aqueous ammonia

solution to quench the reaction. The reaction mixture was precipitated in cold water and

dried under reduced pressure. The resulting quaternized polybutadiene was analyzed by

NMR-spectroscopy.

The NMR signals of the ammonium ion’s N-methyl groups are interfering with the

polybutadiene signals. The methyl group of the n-butyl rest attached to the nitrogen atom

has a distinctive shift in 1H-NMR spectrums and can be used for the monitoring of

conversion from amine to ammonium ion groups (Figure 43). Lorentzian function was used

for peak fitting due to overlapping signals. Elemental analysis is supporting the NMR results.

Figure 43. Signal changes over the course of the reaction

A typical progression of quaternization at room temperature in THF over time is shown in

Figure 44. 5 eq iodomethane relating to the concentration of amine groups were added. 3 eq

at the beginning of the reaction and 2 eq after 5 h. Additionally 2 eq potassium carbonate

were added at the beginning.


70

Figure 44. Conversion of quaternization reaction

The reaction rate is very high at the beginning of reaction and slowing down significantly

with more ammonium ions forming. A reduced solubility with increasing polarity of the

polybutadiene could be the reason. A sterically more hindered center of reaction with

proceeding methylation might be another reason. The introduction of iodomethane after 5 h

seemed to boost the reaction and preventing further deceleration. At high conversions a

distinction of peaks in 1H-NMR spectra is challenging which results in increasing error in

conversion determination. The cyclic tertiary amine groups are most probably slower in

quaternization reaction compared to the secondary amine groups.

5.4.4 Material properties

The resulting quaternized polybutadiene at room temperature is no longer a liquid rubbery

material, but hard and brittle. The polymer structure is resembling polyquaterniums, cationic

polyelectrolytes e.g. used in personal care industry as antistatics and film former.[73] The

cationic charges along the chain are expected to repel themselves and lead to a more

expended rod-like structure compared to the random-coil structure in nonionic

polybutadiene.[109] These conformational changes have a significant impact on the polymer

bulk properties like melt viscosity. The increasing polarity seems also to increase chain-

chain interaction and chain stiffness which results in the increasing glass transition

temperature. The glass transition temperature is increasing to 45 °C (DSC). The significant

0 10 20 30 40 50

0

10

20

30

40

50

60

70

80

90

100

Convers

ion [%

]

time [h]


71

increase in polarity is the reason for increasing solubility in more polar solvents. The product

is soluble in methanol.

Based on conversion of aminolysis and quaternization 20% of initial double bonds are

transferred to an ammonium entity if 30% epoxidated polybutadiene is used. For 42%

epoxidated polybutadiene 25% of initial double bonds are holding an ammonium entity after

quaternization. For Lithium PM4 that would be five to seven ammonium ion entities per

chain, or one ammonium ion every 16-22 carbon atoms in average.

Summary

72

6 Summary

Low molecular weight polybutadienes of molecular weights between 1500 and 26000 g/mol

(Mn) were used in this thesis as base materials. The microstructure of these polybutadienes

showed mainly 1,4 double bonds (80-90%). The polybutadienes were epoxidized using in-

situ formed performic acid in a two-phase system. The polybutadienes were dissolved in

toluene and formic acid was added to the solution at elevated temperatures. Hydrogen

peroxide then was added continuously to the stirred solution. This semi batch process was

scaled up to 10 L and the product properties could be controlled by the chosen reaction

parameters. The degree of epoxidation could be regulated by the chosen temperature,

hydrogen peroxide feed-rate and reaction duration. 1,4 Double bonds showed higher

reactivity towards epoxidation than vinyl pendant groups. The epoxide groups therefore

were almost exclusively introduced on the polymer backbone. The viscosity and glass

transition temperature of the polybutadienes were increased by epoxidation on account of

increasing polarity and stiffness of the chains.

An empirical model was developed to predict the increase of viscosity by epoxidation. The

influence of molecular weight, concentration in toluene, degree of epoxidation and

temperature were investigated by rheological oscillatory shear experiments. Minor

influences of chain entanglements with the higher molecular weight polybutadiene samples

were found. A correction factor was implemented to account for this effect. The viscosity as

function of the molecular weight of the polybutadienes scales with concentration and degree

of epoxidation. The viscosity molecular weight curves were shifted towards higher

viscosities with higher epoxidation ratio or higher concentration. Shift factors were found to

meet these influencing parameters. An ARRHENIUS plot of the zero-shear viscosity at

different temperatures revealed no influence of molecular weight on the flow activation

energy of the polybutadienes. A formula for the prediction of zero shear viscosity of

epoxidized low molecular weight polybutadienes could be developed by combining the shift

factors for molecular weight, concentration, degree of epoxidation and temperature.

The epoxidized polybutadienes were subjected to nucleophilic ring opening by amines. The

epoxide groups located on the polymer backbone showed relatively low reactivity towards

ring opening conditions. Elevated pressures and temperatures as well as Lewis acid catalysis

were necessary for quantitative conversions. The reactions were performed in bulk with the

Summary

73

amine as solvent in high excess and lithium bromide as catalyst. Lithium bromide was

chosen as catalyst because of its low cost and easy removal from the amine modified

polybutadiene product. However, even with catalyst concentrations of 30 mol% quantitative

conversion was achieved only after 24 hours. Apart from this low reactivity there was a

mismatch between epoxide group conversion and aminol formation, hinting towards

occurring side reactions. An intramolecular ring opening of neighboring epoxide groups was

identified as a possible mechanism. The probability of existing neighboring epoxide groups

increases with increasing degree of epoxidation due to the statistical nature of the

epoxidation reaction. A lower maximum degree of amine modification is therefore achieved

with higher degree of epoxidated polybutadienes.

The amine groups in the aminated polybutadiene were alkylated using iodomethane and

potassium carbonate at room temperature. The secondary amine groups were N-

permethylated to get access to quaternary ammonium ions on the polybutadiene backbone.

The introduction of ionic groups to the polymer chain had a significant impact on bulk

properties such as glass transition temperature, which rose above room temperature making

this polyelectrolyte a hard and brittle material. Additionally, solubility in polar solvents like

methanol was increased.

Experimental Part

74

7 Experimental Part

7.1 Materials

Toluene (technical grade) was obtained from VMP Chemiekontor. hydrogen peroxide

(30 wt% aq.) was purchased from VWR Chemicals or Grüssing GmbH Analytika. formic

acid (analytical grade, 99%) was purchased from Grüssing GmbH Analytika, n-butylamin

(99%) was purchased from ABCR GmbH, lithium bromide (>=99%) was purchased from

Sigma-Aldrich, Chloroform-d was purchased from Deutero GmbH. All chemicals were used

without prior purification unless otherwise specified. Four different types of low molecular

weight polybutadienes were used in this study. Lithene ultra© PM4 (Mn = 1500 g/mol) and

Lithene ultra© N4-5000 (Mn = 5000 g/mol) were kindly provided by Synthomer plc. LBR-

307 (Mn = 8000 g/mol) and LBR-305 (Mn = 26000 g/mol) were kindly provided by Kuraray

Europe GmbH. Polybutadienes were used as received.

Table 20: Polybutadienes used in this work with molecular configuration

PBu molecular mass (Mn)

[g/mol]

1,4-DB

[%]

1,2-DB

[%]

cyclic-DB

[%]

cis:trans

ratio7

Lithene PM4 1500 80 19 <1 40:60

Lithene N4-5000 5000 88 12 0 40:60

LBR-307 8000 86 14 0 40:60

LBR-305 26000 91 9 0 40:60

7.2 Nuclear Magnetic Resonance (NMR)

NMR spectra were recorded at room temperature with a Bruker AVANCE 400 MHz or a

Bruker Fourier 300 MHz spectrometer. Chloroform-d was used as solvent.

7.3 Matrix Assisted Laser Desorption (Time of Flight) Mass Spectrometry

(MALDI)

MALDI spectra were recorded on a Bruker UltrafleXtreme Smartbeam II Laser

7 Due to overlapping peaks, the ratios are just approximate values

Experimental Part

75

7.4 Rheometry

Rheometry measurements were done on an ARES G2 Rheometer from TA instruments with

a 60 mm 2° steel cone geometry and peltier plate. The software TRIOS v4.2.1.36612 was

used for measurements and data evaluation.

7.5 Differential Scanning Calorimetry (DSC)

A Mettler Toledo DSC 1 or DSC 821 instrument calibrated with indium was used for thermal

analysis. The weight of samples sealed in an aluminium pan was about 10-15 mg.

First, the samples were cooled from room temperature to -20 °C at -20 K/min, held for 2

min. In a second step, the samples were heated from -20 °C to 60 °C at 10 K/min and held

for 2 min to erase the thermal history. After a second cooling from 60 °C to -120 °C at -20

K/min and second holding for 2 min, the samples were heated from -120 °C to 60 °C at 10

K/min to record heating curves of each sample.

7.6 Gel Permeation Chromatography (GPC)

GPC elugrams were measured on a set up using a precolumn MZ Gel SDplus, 5 µm, 100 Å

from MZ-Analysentechnik, two 5µm Polypore columns from Polymer Laboratories (now

Agilent), an HPLC Pump Intelligent Pump AI12 from Flom, a degasser PLDG 802 from

Polymer Laboratories (now: Agilent), a RI Detector RI 101 from Shodex. The interface hs

2600 and software NTeQGPC V 6.4 version V 1.0.25 used were from hs GmbH.

7.7 Titration of Epoxidized Polybutadienes

Titration of epoxidized polybutadienes for determination of epoxide content was done using

the automatic titration system Titroline alpha plus from SI Analytics. A pH-electrode N4680

eth was used. The end-point determination was done potentiometric with the software

TitriSoft 4.3.

Experimental Part

76

7.8 Epoxidation of Polybutadienes

7.8.1 Reaction Setup

The epoxidation of low molecular weight polybutadienes was performed in various reaction

vessels of glass (250 mL) or steel (1 and 10 L). The double walled cylindrical 250 mL glass

reactor (Büchi AG) was equipped with an anchor stirrer, a thermostat and a syringe pump.

The double walled cylindrical 1 L and double walled conical 10 L steel reactor (Juchheim

Laborgeräte GmbH) were equipped with an anchor stirrer, a thermostat and a double piston

pump (Pharmacia LKB P-500) for uninterrupted feeding of hydrogen peroxide. The

conversion showed a dependence on the reactor material: The first couple of reactions in

steel reactors showed lower conversions, i.e. before the surface was deactivated. This may

be related to metallic ions accelerating the decomposition of hydrogen peroxide.[110] After

each experiment the vessel was extracted three times with toluene for 30 min at the reaction

temperature.

7.8.2 General Procedure of Epoxidation with in-situ Formed Performic Acid

PBu was dissolved in toluene to a concentration of 10-15 wt%. The solution was transferred

to the reactor and heated to reaction temperature. Formic acid (FA) was added. Hydrogen

peroxide (30 wt%) was subsequently added to the stirred mixture, either in one batch or in

semi-batch mode. After the reaction, the organic phase of the final reaction mixture was

phase separated from the aqueous phase, and subsequently washed twice with a saturated

NaHCO3-solution and with a saturated NaCl-solution to remove residual peroxide and acid.

The organic phase was then dried over Na2SO4 and the toluene was removed under reduced

Figure 45. Reaction set-up of the semi-batch process

Experimental Part

77

pressure at 60 °C. The epoxidized polybutadienes were obtained in yields between 90 and

98% based on the applied PBu.

7.8.3 Degree of Epoxidation

The degree of epoxidation was conveniently determined from the 1H-NMR spectra in CDCl3.

The degree of epoxidation was calculated using equation 12. The integrals refer to the signals

indicated by the chemical shift δ in ppm.

( 2.7 2.9 )[%] 100%

( 2.7 2.9 ) ( 5.0 ) ( 5.4 5.6 )

ppmX

ppm ppm ppm

(12)

The signals of the 1,2- and 1,4-double bonds ( = 5.0 − 5.6 broaden as a result of the

epoxidation, the integrals however can be obtained without additional effort. Additionally,

elemental analysis was performed to determine the degree of double bond epoxidation. The

results of elemental analysis are in accordance with the 1H-NMR results (Table 21).

Table 21. Comparison of results of conversion by 1H-NMR and elemental analysis.

conversion by 1H-NMR [%] conversion by elemental analysis [%]

10 12

21 21

30 32

37 37

39 38

42 42

The progress of the reaction was monitored from samples taken from the reacting mixture.

The stirring was stopped, a sample of about 2 mL was taken from the reaction mixture with

a pipette and transferred to a glass vial. It was treated two times with 1 mL of NaHCO3

solution and once with 1 mL of NaCl solution. The organic phase was transferred to another

glass vial and the toluene removed under reduced pressure.

Experimental Part

78

7.8.4 Titration of Epoxide Groups

The titration was based on the method of JUNG and KLEEBERG.[78] The epoxidized PBu

sample was put into a 50 mL Erlenmeyer flask. The sample mass added was calculated by 1H-NMR results, that 1.5 mmol epoxide groups were present in the sample. 5 mL butanone

was added and slewed until the PBu sample was dissolved. 10 mL of a solution (8 mL

concentrated hydrochloric acid in 300 mL butanone) was added and the closed flask stored

for 30 min. 5 mL of demineralized water was added prior to titration to dissolve potassium

chloride formed during reaction. The titration was conducted using 0.1 M ethanol solution

of potassium hydroxide.

A blank value was determined once every day of measurement. The titer potassium

hydroxide solution was determined using 200 to 300 mg potassium hydrogen phthalate in

boiled demineralized water.

The EP-value was calculated using the following equation:

= ∙ ∙∙ (13)

Vblank and Vsample are the volumes of potassium hydroxide solution used until titration end

point of the blank value and the sample titration respectively. c(KOH) describes the

concentration of potassium hydroxide solution and tKOH the titer of the same. The samples

mass is given by msample. The EP-value describes the molar amount of epoxide per 100 g of

epoxidized PBu, which can be calculated into a relative epoxidation ratio with the knowledge

of the polymers molecular mass.

Experimental Part

79

7.9 Rheological Characterization of Epoxidized Polybutadienes

7.9.1 Materials

Toluene was destilled and stored over molecular sieves (4 Å) prior to use.

7.10 Oscillatory Rheology measurements

A strain sweep measurement from 0.1% to 1% or 10% was done for every sample type to

determine the linear viscoelastic range of the material under the given conditions. The strain

for the frequency measurements was then set to a value well in the linear region of the

complex oscillatory viscosity.

For the toluene diluted samples a solvent trap cap was used to prevent solvent evaporation

during the measurement.

The time-temperature-superposition (TTS) mastercurves were calculated using the TRIOS

software and with the measurement at 25 °C as reference curve.

Experimental Part

80

7.11 Aminolysis of Epoxidized Polybutadienes

7.11.1 Process

The ring opening of epoxidized 1,4-polybutadiene is much slower compared to epoxidized

1,2-polybutadiene (vide supra). High temperatures are therefore needed, which exceed the

boiling point of n-butylamine (b.p.: 78 °C). These reactions are thus conducted in glass

autoclave vessels sealed by a PTFE cap and PTFE sealing tape (35 mL, max. 10 bar) at

temperatures above boiling point. The vessels were heated in a glass beaker filled with

sunflower oil. A Heidolph MR 3002 control heating plate was used as heating device.

1 or 2 g of the epoxidized polybutadiene was dissolved under stirring in 20 mL of butyl

amine, lithium bromide was added, the vessel was sealed tight and heated to the reaction

temperature.

After the reaction time the vessel was allowed to cool, the excess amine was evaporated

under reduced pressure and the resulting high viscous product was precipitated out of THF

in water to remove residual free n-butylamine. The product was received as a highly viscous,

slightly yellow liquid.

7.11.2 Degree of Aminolysis

The decrease of epoxy content and increase of amine functionalization was monitored by 1H-NMR and 13C-NMR I-Gated measurements. In 1H-NMR spectroscopy, the signals at

2.7 ppm and 2.9 ppm are attributed to the trans and cis epoxy groups respectively. In 13C-

NMR I-Gated measurements the signals at 56.7 ppm and 58.3 ppm are attributed to the cis

and trans epoxy groups respectively (Figure 46). The analysis of the amine conversion is

illustrated exemplarily for the aminolysis with n-butylamine, where the signal of the terminal

methyl group is shown at 0.92 ppm in 1H-NMR measurements. The signals of the remaining

methylene groups coincide with the signals of the polybutadiene chain. In 13C-NMR I-Gated

measurements the signal at 14.0 ppm is attributed to the terminal methyl group of butylamine

and the signal at 20.5 ppm to the adjacent methylene group (Figure 46).

Experimental Part

81

Figure 46: 13C-NMR I-Gated spectra of the epoxidized sample (a) and the aminated sample (b).

The integrals of the peaks in 13C-NMR I-Gated spectra are only reliable with long relaxation

times. Because of the resulting very long measuring times the degree of aminolysis was

determined by 1H-NMR spectra and elemental analysis measurements. However qualitative

results were drawn from the 13C-NMR I-Gated spectra. A 13C-NMR I-Gated measurement

with a relaxation time of 80 s showed good accordance with the results of 1H-NMR and

elemental analysis (Table 16).

Experimental Part

82

Figure 47: 1H-NMR spectrum of the epoxidized sample

Figure 48: 1H-NMR spectrum of the aminated sample

The Degree of aminolysis relative to the initial double bonds was therefore determined using 1H-NMR (equation 14).

The Integrals were normalized to the integral of the t signal, which is visible without

overlapping in epoxidated polybutadiene as well as in aminated polybutadiene and doesn’t

change during aminolysis reaction.

Experimental Part

83

The integrals of n, o, f and g in the spectrum of the epoxidated polybutadiene precurser

(EBR) were divided by two, since two protons represent one epoxide group. The integral of

x in the spectrum of the aminated polybutadiene product (ABR) was divided by three, since

three protons represent one amine group. The integrals of b, c, j, k, t in the spectra of

epoxidized as well as aminated polybutadiene were divided by two, since two protons

represent one double bond. The corrected integrals were used in equation (14.

Σ , , , , , , , , , = , Σ , , , , , , =

(14)

% = + − + + , , , ∗ 100

The difference of EBR and ABR equals the integrals of epoxide groups not opened by n-

butylamine, representing either unopened epoxide groups or epoxide groups opened by a

neighboring aminol group forming a cyclic ether or cyclic amine (see chapter 5.3.4). The

formation of cycles does open an epoxide group without the addition of n-butylamine.

Experimental Part

84

7.12 Quaternization of aminated polybutadienes

7.12.1 Process

In a glass flask equipped with a magnetic stirrer 1-10 g aminated polybutadiene were

dissolved in THF (10-15 times by weight) under stirring at room temperature. 2-3 eq

potassium carbonate, relative to the concentration of amine groups, were added under

stirring. The first half of 5-10 eq iodomethane, relative to the concentration of amine groups,

were added to the resulting suspension to start the reaction. The flask was sealed, to avoid

evaporation of iodomethane. During the reaction a fine colorless precipitation was visible,

which was distinguishable from the coarse potassium carbonate and most likely was

potassium iodide. After a certain time (usually 24 h) the second half of iodomethane was

added.

After 72 h the reaction was quenched by adding 50 mL of aqueous ammonia (30%). The

resulting mixture was precipitated in cold water, filtered, washed with water and dried under

reduced pressure. The product was received as a hard, slightly brown foam.

7.12.2 Degree of Quaternization

The degree of quaternization was monitored using 1H-NMR spectroscopy. The triplet at

0.92 ppm, attributed to the methyl group of bound n-butylamine, shifts downfield with the

introduction of quaternary ammonium ions (Figure 49). The remaining methylene groups of

bound n-butylamine shift as well but coincide with the rubber signals of polybutadiene. The

shifted triplet has still some overlapping with the triplet of non-ionized amines methyl group,

which is why a lorentzian fit was used to distinguish both peaks (Figure 50).

Experimental Part

85

Figure 49. Assignement of the two overlapping triplets

Figure 50. Lorentzian fit (red line) of the two overlapping triplets

Experimental Part

86

The conversion was calculated using equation (15):

X% = 0.90, 0.92, 0.95 0.90, 0.92, 0.95 + 0.83, 0.85, 0.87 ∗ 100 (15)

Safety Data

87

8 Safety Data

Table 22. Hazard statements according to GHS[108]

Substance GHS pictograms Hazard Sentences Precaution Sentences

1,2-Epoxybutane

Danger

H225,

H302+H312+H332,

H315, H319, H335,

H351, H412

P21, P261, P273, P280,

P305+P351+P338,

P308+P313, P403+P235

2-(Dimethylamino)-

ethanol

Danger

H226, H331, H312,

H302, H314, H335

P210, P280,

P303+P361+P353,

P305+P351+P338,

P312, P405

2,3-Epoxybutan

Danger

H225, H315, H319,

H335

P210, P261,

P305+P351+P338

2-Amino-2-

methylpropan-1-ol

Danger

H315, H318, H412 P273, P280,

P305+P351+P338

Aqueous ammonia

(30%)

Danger

H290, H314, H335,

H400

P260, P273, P280,

P301+P330+P331,

P303+P361+P353,

P305+P351+P338

Butanone

Danger

H225, H319, H336 P210,

P305+P351+P338,

P403+P233

Chloroform-d

Danger

H302, H331, H315,

H319, H351, H361d,

H372

P201, P260, P264,

P280,

P304+P340+P311,

P403+P233

Benzil

Caution

H315, H319 P302+P352,

P305+P351+P338

Safety Data

88


Dimethylsulfate

Danger

H301, H314, H317,

H330, H341, H350

P201, P280,

P301+P330+P331,

P302+P352,

P304+P340,

P305+P351+P338,

P308+P310

Formic acid

Danger

H226, H302, H314,

H331

P210, P280,

P303+P361+P353,

P304+P340+P310,

P305+P351+P338,

P403+P233

Hydrochloric acid

aq. (36%)

Danger

H290, H314, H335 P260, P280,

P303+P361+P353,

P304+P340+P310,

P305+P351+P338

Hydrogen peroxide

aq. (30%)

Danger

H318, H412 P280,

P305+P351+P338+P310

Iodomethane

Danger

H301+H331, H312,

H315, H319, H335,

H351, H410

P273, P302+P352,

P304+P340,

P305+P351+P338,

P308+P310

LBR-305 No hazardous substance

LBR-307 No hazardous substance

Lithene ultra® N4-

5000 No hazardous substance

Lithene ultra® PM4 No hazardous substance

Lithium bromide

Caution

H302, H315, H317,

H319

P280, P305+P351+P338

Safety Data

89


Methanol

Danger

H225, H331, H311,

H301, H370

P210, P233, P280,

P302+P352,

P304+P340,

P308+P310, P403+P235

n-Butylamine

Danger

H225, H302,

H311+H331, H314

P210, P233, P240,

P280,

P301+P330+P331,

P302+P352,

P305+P351+P338,

P308+P310, P403+P235

Potassium

carbonate

Caution

H315, H319, H335 P302+P352,

P305+P351+P338

Potassium

hydrogen phthalate No hazardous substance

Potassium

hydroxide 0.1 N in

ethanol

Danger

H225, H315, H319 P210, P240,

P302+P352,

P305+P351+P338,

P403+P233

Sodium

bicarbonate No hazardous substance

Sodium chloride No hazardous substance

Tetrahydrofurane

Danger

H225, H302, H319,

H335, H351

P210, P280,

P301+P312+P330,

P305+P351+P338,

P370+P378, P403+P235

Toluene

Danger

H225, H304, H315,

H336, H361d, H373

P210, P240,

P301+P310+P330,

P302+P352, P314,

P403+P233

Safety Data

90

CMR substances used are listed in Table 23.

Table 23. CMR substances used.[111]

CAS-No. Name Procedure Quantity Catagory

77-78-1 Dimethylsulfate Quarternization ~200 mL Carc. 1B

106-88-7 1,2-Epoxybutane Aminolysis ~14 mL Carc. 1B

67-66-3 Trichloromethane-d NMR-Measurements ~200 mL Carc. 1B

Bibliography

91

9 Bibliography

[1] H.-G. Elias in Makromoleküle (Ed.: H.-G. Elias), 1999.

[2] D. Zuchowska, Polymer 1980, 21, 514–520.

[3] M. M. Jacobi, C. K. Santin, M. E. Viganico, R. H. Schuster, Kautschuk Gummi

Kunststoffe 2004, 3/2004, 82–89.

[4] R. J. Salinas, R. B. Petrovic, Synthetic Rubber Overview, Goa, India, 2016.

[5] F. Röthemeyer, F. Sommer (Eds.) Kautschuk-Technologie. Werkstoffe - Verarbeitung -

Produkte, Hanser, München [u.a.], 2006.

[6] S. V. Lebedev, Zh. Russ. Fiz.Khim. Ova. Chast. 1910, 42, 949.

[7] C. Harries, Justus Liebigs Ann. Chem. 1911, 383, 157–227.

[8] F. E. Mathews, E. H. Strange, GB 24,790, 1910.

[9] I.G. Farbenfabriken, FR 668113, 1928.

[10] Phillips Petroleum Comp., GB 848065, 1956.

[11] M. D. Lechner, K. Gehrke, E. Nordmeier, Makromolekulare Chemie. Ein Lehrbuch

für Chemiker, Physiker, Materialwissenschaftler und Verfahrenstechniker, Birkhäuser,

Basel, 2010.

[12] F. E. Matthews, E. H. Strange, GB24.700, 1910.

[13] P. Cossee, Tetrahedron Letters 1960, 1, 12–16.

[14] P. Cossee, Journal of Catalysis 1964, 3, 80–88.

[15] P. Cossee, Recl. Trav. Chim. Pays-Bas 1966, 85, 1151–1160.

[16] E. Arlman, Journal of Catalysis 1964, 3, 89–98.

[17] E. Arlman, Journal of Catalysis 1964, 3, 99–104.

Bibliography

92

[18] C. Costabile, G. Milano, L. Cavallo, G. Guerra, Macromolecules 2001, 34, 7952–

7960.

[19] Pigment & Resin Technology 1977, 6, 11–15.

[20] E. Piorkowska, A. S. Argon, R. E. Cohen, Polymer 1993, 34, 4435–4444.

[21] H. R. Brown, A. S. Argon, R. E. Cohen, O. S. Gebizlioglu, E. J. Kramer,

Macromolecules 1989, 22, 1002–1004.

[22] H. Fischer, B. Wegemund, W. Gress, M. Gorzinksi, DE3442200 A1, 1984.

[23] S. Ichimura, Y. Tsuzuki, K. Hino, US5965255, 1997.

[24] A. Konietzny, H.-D. Zagefka, K. Rombusch, H.-J. Bax, DE3233949 A1, 1982.

[25] D. C. Edwards, K. Sato, Rubber Chemistry and Technology 1980, 53, 66–79.

[26] Y. Arai, US4594388, 1984.

[27] J. Y. Choi, J. K. Lee, Y. You, W. H. Park, Fibers Polym 2003, 4, 195–198.

[28] G. Heublein, B. Heublein, P. Hortschansky, H. Meissner, H. Schutz, Journal of

Macromolecular Science: Part A - Chemistry 1988, 25, 183–200.

[29] J. W. Holubka, US4486571, 1983.

[30] M. Soga, Y. Takase, M. Ishikawa, M. Nakazawa, M. Nakayama, US4074034,

1975.

[31] Y. Tsuchiya, K. Ito, K. Hagihara, H. Sakamoto, Y. Otsuki, US4730010, 1986.

[32] H. Yasuno, T. Yoshinaga, T. Nakano, US4250007, 1979.

[33] A. N. Bibi, D. A. Boscott, T. Butt, R. S. Lehrle, European Polymer Journal 1988,

24, 1127–1131.

[34] P. Bussi, H. Ishida, J. Polym. Sci. B Polym. Phys. 1994, 32, 647–657.

[35] A. G. Margaritis, N. K. Kalfoglou, European Polymer Journal 1988, 24, 1043–

1047.

Bibliography

93

[36] J.-P. Dilcher, H. Jürgens, G. A. Luinstra in Advances in Polymer Science, Vol. 269

(Ed.: P. Theato), Springer International Publishing, Cham (Switzerland), 2015.

[37] Q. Zhou, S. Jie, B.-G. Li, Polymer 2015, 67, 208–215.

[38] F. P. Greenspan, R. E. Light, US2829135, 1955.

[39] C. Wheelock, Ind. Eng. Chem. 1958, 50, 299–304.

[40] W. Kirchhof, W. Stumpf, B. Schleimer, US3253000, 1962.

[41] H. S. Makowski, M. Lynn, D. H. Rotenberg, Journal of Macromolecular Science:

Part A - Chemistry 1970, 4, 1563–1597.

[42] T. Hirai, Y. Hatano, S. Nonogaki, J. Electrochem. Soc. 1971, 118, 669.

[43] El Fayoumi, Ali A. Z., Journal of Thermal Analysis 1982, 23, 135–141.

[44] W.-K. Huang, G.-H. Hsiue, W.-H. Hou, J. Polym. Sci. A Polym. Chem. 1988, 26,

1867–1883.

[45] R. Siegmeier, A. Grund, G. Prescher, U. Brandt, 4851556, 1988.

[46] O. Hayashi, K. Kimura, Y. Ooi, H. Ueno, Kobunshi Ronbunshu 1980, 37, 327–334.

[47] O. Hayashi, T. Takahashi, H. Kurihara, H. Ueno, Kobunshi Ronbunshu 1980, 37,

195–198.

[48] K. Maenz, H. Schütz, D. Stadermann, European Polymer Journal 1993, 29, 855–

861.

[49] S. Roy, Namboodri, C. S. S., B. R. Maiti, B. R. Gupta, Polym. Eng. Sci. 1993, 33,

92–96.

[50] Y. Kurusu, Y. Masuyama, M. Miyamoto, Polym J 1994, 26, 1163–1169.

[51] M. M. Jacobi, C. P. Neto, C. G. Schneider, Rocha, T. L. A. C., R. H. Schuster,

Kautschuk Gummi Kunststoffe 2002, 11/2002, 590–595.

[52] K. Maenz, M. Möllhoff, D. Stadermann, Acta Polym. 1996, 47, 208–213.

Bibliography

94

[53] A. Iraqi, S. Seth, C. A. Vincent, D. J. Cole-Hamilton, M. D. Watkinson, I. M.

Graham, D. Jeffrey, J. Mater. Chem. 1992, 2, 1057.

[54] M. P. Achalpurkar, U. K. Kharul, H. R. Lohokare, P. B. Karadkar, Sep. Purif.

Technol. 2007, 57, 304–313.

[55] D. Zuchowska, Makromol. Chem. Rapid Commun. 1981, 2, 135–138.

[56] A. K. Chakraborti, S. Rudrawar, A. Kondaskar, European Journal of Organic

Chemistry 2004, 2004, 3597–3600.

[57] T. Ollevier, G. Lavie-Compin, Tetrahedron Letters 2002, 43, 7891–7893.

[58] A. Solladié-Cavallo, P. Lupattelli, C. Bonini, The Journal of organic chemistry

2005, 70, 1605–1611.

[59] A. T. Placzek, J. L. Donelson, R. Trivedi, R. A. Gibbs, S. K. De, Tetrahedron

Letters 2005, 46, 9029–9034.

[60] M. Fujiwara, M. Imada, A. Baba, H. Matsuda, Tetrahedron Letters 1989, 30, 739–

742.

[61] M. Meguro, N. Asao, Y. Yamamoto, J. Chem. Soc., Perkin Trans. 1 1994, 2597.

[62] J. Augé, F. Leroy, Tetrahedron Lett 1996, 37, 7715–7716.

[63] G. R. Krishnan, K. Sreekumar, Polymer 2008, 49, 5233–5240.

[64] C. M. Kleiner, P. R. Schreiner, Chem. Commun. 2006, 4315.

[65] M. Vijender, P. Kishore, P. Narender, B. Satyanarayana, Journal of Molecular

Catalysis A: Chemical 2007, 266, 290–293.

[66] H. Chen, H. Zhang, D. Thor, R. Rahimian, X. Guo, Eur. J. Med. Chem. 2012, 52,

159–172.

Bibliography

95

[67] V. M. Dem’yanovich, I. N. Shishkina, A. A. Kuznetsova, K. A. Potekhin, A. V.

Chesnova, Russ. J. Org. Chem. (Russian Journal of Organic Chemistry) 2006, 42,

986–989.

[68] C. Zhang, H. Ito, Y. Maeda, N. Shirai, S.-i. Ikeda, Y. Sato, J. Org. Chem. 1999, 64,

581–586.

[69] T. Hunt, H. C. Atherton-Watson, S. P. Collingwood, K. J. Coote, S. Czarnecki, H.

Danahay, C. Howsham, P. Hunt, D. Paisley, A. Young, Bioorganic & Medicinal

Chemistry Letters 2012, 22, 2877–2879.

[70] C. Roberts, W. E. Lindsell, I. Soutar, Brit. Poly. J. 1990, 23, 55–62.

[71] J. Hazziza–Laskar, N. Nurdin, G. Helary, G. Sauvet, J. Appl. Polym. Sci. 1993, 50,

651–662.

[72] K. Dorfner, Ion Exchangers, De Gruyter, 2011.

[73] R. Schueller, P. Romanowski, Conditioning Agents for Hair and Skin, Taylor &

Francis, 1999.

[74] N. Phan-Thien, Understanding Viscoelasticity. An Introduction to Rheology,

Springer, Berlin, Heidelberg, 2013.

[75] TA instruments -, "Understanding Rheology of Thermoplastic Polymers".

[76] W. P. Cox, E. H. Merz, J. Polym. Sci. 1958, 28, 619–622.

[77] T. Mezger, Das Rheologie-Handbuch. Für Anwender von Rotations- und

Oszillations-Rheometern, Vincentz Network, Hannover, 2012.

[78] G. Jung, W. Kleeberg, Kunststoffe 1961, 51, 714–715.

[79] B. Kona, M. St. Weidner, J. F. Friedrich, International Journal of Polymer Analysis

and Characterization 2005, 10, 85–108.

Bibliography

96

[80] N. V. Bac, L. Terlemezyan, M. Mihailov, J. Appl. Polym. Sci. 1991, 42, 2965–

2973.

[81] S. Gnecco, A. Pooley, M. Krause, Polymer Bulletin 1996, 37, 609–615.

[82] S.-M. Wang, R. C.-C. Tsiang, J. Polym. Sci. A Polym. Chem. 1996, 34, 1483–1491.

[83] M. I. Abdullin, A. A. Basyrov, O. S. Kukovinets, A. B. Glazyrin, G. I.

Khamidullina, Polym. Sci. Ser. B 2013, 55, 349–354.

[84] N. E. Gordon, Dissertation, Johns Hopkins University, Baltimore, Maryland, 1917.

[85] Kolossowsky, Megenine, Bull. Soc. Chim. Fr. 1932, 51, 1000–1004.

[86] Atherton Seidell, Solubilities of Organic Compounds. A compilation of Quantitative

Solubility Data from the Periodical Literature, D. Van Nostrand Company, Inc., New

York, 1941.

[87] X. SUN, X. ZHAO, W. DU, D. LIU, Chinese Journal of Chemical Engineering

2011, 19, 964–971.

[88] J. P. Dilcher, Dissertation, Universität Hamburg, Hamburg, Unveröffentlicht.

[89] P. D. Filippis, M. Scarsella, N. Verdone, Ind. Eng. Chem. Res. 2009, 48, 1372–

1375.

[90] J. K. Jackson, M. E. de Rosa, H. H. Winter, Macromolecules 1994, 27, 2426–2431.

[91] M. Ballauff, Angewandte Chemie 1989, 101, 261–276.

[92] G. C. Berry, T. G. Fox, Adv. Polymer Sci. 1968, 5, 261–357.

[93] B. L. Hager, G. C. Berry, J. Polym. Sci. Polym. Phys. Ed. 1982, 20, 911–928.

[94] R. W. Rendell, K. L. Ngai, G. B. McKenna, Macromolecules 1987, 20, 2250–2256.

[95] Y. Takahashi, Y. Isono, I. Noda, M. Nagasawa, Macromolecules 1985, 18, 1002–

1008.

Bibliography

97

[96] De Risi, F. R., L. D'Ilario, A. Martinelli, J. Polym. Sci. A Polym. Chem. 2004, 42,

3082–3090.

[97] I. T. Smith, Polymer 1961, 2, 95–108.

[98] Y. Kurusu, Polym. Adv. Technol. 1996, 7, 67–72.

[99] B. A. Rozenberg in Advances in Polymer Science, Vol. 75 (Eds.: K. Dušek, J. P.

Bell), Springer-Verlag, Berlin, New York, 1986.

[100] P. K. Sundaram, M. M. Sharma, Bull. Chem. Soc. Jpn. 1969, 42, 3141–3147.

[101] I. R. Gelling, Rubber Chemistry and Technology 1985, 58, 86–96.

[102] I. R. Gelling, Prog. Rubb. & Plast Tech. 1991, 7, 271–297.

[103] A. Challioui, D. Derouet, A. Oulmidi, J. C. Brosse, Polym. Int. 2004, 53, 1052–

1059.

[104] S. Roth, S. Göhler, H. Cheng, C. B. W. Stark, Eur. J. Org. Chem. 2005, 2005,

4109–4118.

[105] S. Göhler, S. Roth, H. Cheng, H. Göksel, A. Rupp, L. Haustedt, C. Stark, Synthesis

2007, 2007, 2751–2754.

[106] G. A. Sulikowski, M. M. Sulikowski, M. H. Haukaas, B. Moon, Iodomethane,

2005, American Cancer Society, zu finden unter

https://onlinelibrary.wiley.com/doi/full/10.1002/047084289X.ri029m.pub2.

[107] H. Z. Sommer, L. L. Jackson, J. Org. Chem. 1970, 35, 1558–1562.

[108] IFA, "GESTIS-Stoffdatenbank", zu finden unter

http://gestis.itrust.de/nxt/gateway.dll?f=templates&fn=default.htm&vid=gestisdeu:sdb

deu.

[109] J. E. Mark, J. Am. Chem. Soc. 1966, 88, 4354–4359.

[110] C. Walling, A. Goosen, J. Am. Chem. Soc. 1973, 95, 2987–2991.

Bibliography

98

[111] Deutsche Gesetzliche Unfallversicherung, "Liste der krebserzeugenden,

keimzellmutagenen und reproduktionstoxischen Stoffe (KMR-Stoffe)", zu finden unter

https://publikationen.dguv.de/dguv/pdf/10002/kmr_oktober_2018.pdf, 2018.

Acknowledgements

99

10 Acknowledgements

In this chapter I like to thank a number of people who have accompanied and supported me

during my PhD. This part is written in German.

Zunächst möchte ich mich bei Herrn Professor Dr. Gerrit A. Luinstra bedanken für die

Aufnahme in seinen Arbeitskreis, die fachliche Betreuung und die Ermöglichung der

Promotion trotz meiner kreativen Schaffenspause nach der Diplomarbeit.

Herrn Professor Dr. Hans-Ulrich Moritz möchte ich für die Übernahme des Zweitgutachtens,

sowie informative und unterhaltsame Vorlesungen danken.

Kathrin Rehmke und Stefan Bleck danke ich für die Messung der zahlreichen GPC und DSC

Proben. Jens Pagel möchte ich für die kompetente und unkomplizierte Hilfe bei Reaktorbau

und anderen technischen Problemen danken. Holger Stockhusen danke ich für die Hilfe bei

Problemstellungen mit der Elektrik. Kathleen Pruntsch danke ich für die reibungslose

Abwicklung von Bestellungen und Laborbedarfsanforderungen, sowie ihr fröhlich direktes

Auftreten.

Dem gesamten AK Luinstra danke ich für die entspannte, lockere Atmosphäre, die Hilfe bei

fachlichen und organisatorischen Problemen, sowie die zahllosen Grillabende, Bierrunden,

Exkursionen, private Feiern, AK Fahrten und sportlichen Kampfhandlungen digitaler und

analoger Natur. Meinen Labormitbewohnern Claudi, Franzi, Daniel, Helena und Karen

danke ich für die kurzweilige und motivierende Arbeitsatmosphäre, sowie Gesprächen

innerhalb und abseits der Chemie. Daniel danke ich außerdem für die abwechslungsreiche

Musikauswahl, die geduldige theoretische und praktische Unterstützung (vor allem in der

Rheologie) und den hohen Entertainment Wert besonders nach ein paar Bierchen.

Herausheben möchte ich außerdem Jan für die Unterstützung bei Epoxidierung, Titration,

etc. und für die Fussballdiskussionen und Lars für Filmabende und Rain in Africa.

Ich danke meinen Praktikanten und Bachelorstudenten für die Unterstützung bei der

Synthesearbeit und besonders Stefanie Schmidt, Fabian Wenzel und Fabian Ratzke für ihren

engagierten Einsatz.

Acknowledgements

100

André, Dirk, Martin, Marcel und Matthias, danke ich für die Standhaftigkeit unter digitalem

Artilleriefeuer und Beschuss, die Stammtischrunden, Operation „Mustache“,

Junggesellenabschiede, Hochzeiten, Feiern, und und und.

Außerhalb der Uni möchte ich besonders Benni danken für seine vielen Jahre als Freund,

Kumpel und WG-Mitbewohner. Zusammen mit Jan und Tim gab es so schön viel Ablenkung

vom manchmal stressigen Uni-Alltag beim ein oder anderen Bier - egal wo (hauptsache

warm und Musik). Danke euch Jungs!

Meinen Schwiegereltern Elke und Karl-Heinz möchte ich danken für ihre Unterstützung auf

ganz vielen Ebenen und die sehr herzliche Aufnahme in die Familie.

Meinen Eltern gilt ganz besonderer Dank für ihre Unterstützung, ihren Stolz und ihre

Geduld. Vielen Dank dafür, dass ich immer auf euch zählen kann!

Meinen Schwestern Lene und Inga danke ich für lustige, gemütliche und spannende Tage,

Abende und Gespräche.

Mein Größter Dank gilt meiner Frau Ulrike, mit der ich nun schon über 13 Jahre mein Leben

teilen darf und seit über zwei Jahren eine wundervolle Tochter, sowie seit kurzem einen

aufgeweckten, kleinen Sohn habe. Ich danke Dir, dass Du mich motivierst, mir den Rücken

stärkst, mich ausgleichst, mich manchmal wieder auf den Weg bringst, Dich nicht verbiegst,

Dir treu bleibst und Dir nicht für Blödeleien zu schade bist!

Declaration of Oath

101

11 Declaration of Oath

I hereby declare on oath, that I have written the present dissertation by my own and have not

used other than the acknowledged resources and aids. I hereby declare that I have not

previously applied or pursued for a doctorate (Ph.D. studies).

Post Modification of Liquid Polybutadienes and Their ...

Documents

Transcript of Post Modification of Liquid Polybutadienes and Their ...